微软&领英：生成式 AI 职业认证课程考试题库

在迈向专业认证的道路上，合理利用题库是备考的智慧选择。本文将深入浅出地解析如何高效利用题库，辅以实际案例，帮助大家把握学习的重点，有效提升备考效率。

获取更多AI资料

扫描下方二维码，关注公众号，回复关键字：ai，获取更多AI资料或者完整题库。

考试题库(答案仅供参考)

第 1 题

Your company wants to create a smartphone application that identifies plants using thephone’s camera. The company purchases millions of digital images of plants labeled withthe species names. You use this initial batch of images to train your artificial neuralnetwork, What type of machine learning are you using for your network?

• reinforcement learning
• self-supervised learning
• unsupervised machine learning
• supervised learning

您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 强化学习
• 自我监督学习
• 无监督机器学习
• 监督学习

在监督学习中，模型是使用标记数据进行训练的，这意味着数据集中的每个示例都与正确的输出配对。该模型从这些数据中学习，然后利用这种学习对新的、未见过的数据进行预测或推断。在这种情况下，标签是植物的物种名称。

第 2 题

A national newspaper reporter is writing a story on generative Al. As part of the story, theychat for hours with a new online generative chatbot. A few hours into the conversation, thechatbot tries to convince the reporter to leave his partner. The chatbot company said theydon’t know why it gave these responses and will limit conversations to 30 minutes. Whatmight be one of the biggest ethical challenges with this system?You work in the marketingdepartment for a large company, and you’d like to create a weekly opinion letter usingChatGPT to give your take on the top news in your industry. You create a few test postsand you notice that ChatGPT is getting the dates wrong and is mixing up the CEos ofdifferent companies. Why are you running into this challenge?Your company wants to usegenerative Al to come up with new pharmaceuticals. This system will analyze all existingchemical compounds and try to develop new compounds based on the success of some ofyour current pharmaceuticals. This system will require a lot of custom programming andaccess to your proprietary data sets. What type of generative Al system might work best?Your company wants to create a smartphone application that identifies plants using thephone’s camera. The company purchases millions of digital images of plants labeled withthe species names. You use this initial batch of images to train your artificial neuralnetwork. What type of machine learning are you using for your network?

• There isn’t enough transparency into how the chatbot is responding.
• The system has access to tremendous amounts of data, so it can offer hard but truthfuladvice.
• The system offered personal advice too soon in the conversation.
• Chatbots shouldn’t offer marital advice.

一位全国性报纸记者正在撰写一篇关于生成性阿尔的报道。作为故事的一部分，他们与新的在线生成聊天机器人聊了几个小时。谈话几个小时后，聊天机器人试图说服记者离开他的搭档。该聊天机器人公司表示，他们不知道为什么会给出这些回复，并将把对话时间限制在 30 分钟内。该系统最大的道德挑战之一可能是什么？您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周意见书，以表达您对行业头条新闻的看法。您创建了一些测试帖子，您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要利用再生人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前的一些药物的成功开发新的化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 聊天机器人的响应方式缺乏足够的透明度。
• 该系统可以访问大量数据，因此可以提供艰难但真实的建议。
• 系统在对话中过早地提供了个人建议。
• 聊天机器人不应该提供婚姻建议。

上述聊天机器人系统面临的最大道德挑战之一是聊天机器人的响应方式缺乏足够的透明度。如果聊天机器人开始提供个人建议并试图说服用户做出重大的生活决定，但没有明确解释为什么这样做，就会引起人们对系统决策过程以及操纵或意外结果的可能性的担忧。透明度对于人工智能系统至关重要，可以确保用户了解系统如何做出决策并在其交互中建立信任。

第 3 题

You work for a large financial institution that would like to offer immediate approval forloan applications. Your team has identified four predictors about whether someone will bea good loan candidate: income, credit score, employment, and debt. You develop a systemthat will look at each predictor independently and then come up with an overall score.What machine learning algorithm are you using?You work in the marketing department fora large company, and you’d like to create a weekly opinion letter using ChatGPT to giveyour take on the top news in your industry, You create a few test posts, and you notice thatChatGPT is getting the dates wrong and is mixing up the CEOs of different companies. Whyare you running into this challenge?Your company wants to use generative Al to come upwith new pharmaceuticals. This system will analyze all existing chemical compounds andtry to develop new compounds based on the success of some of your currentpharmaceuticals. This system will require a lot of custom programming and access to yourproprietary data sets. What type of generative Al system might work best?Your companywants to create a smartphone application that identifies plants using the phone’s camera.The company purchases millions of digital images of plants labeled with the species names.You use this initial batch of images to train your artificial neural network, What type ofmachine learning are you using for your network?

• K-means clustering
• Naive Bayes
• Linear regression
• K-nearest neighbor

您在一家大型金融机构工作，希望立即批准申请。您的团队已经确定了关于某人是否适合贷款的四个预测因素：收入、信用评分、就业和债务，您开发了一个系统，该系统将独立查看每个预测因素，然后得出总体分数。您使用什么机器学习算法使用？您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周意见信，以表达您对行业热门新闻的看法。您创建了一些测试帖子，您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要使用生成式人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• K-均值聚类
• 朴素贝叶斯
• 线性回归
• K-最近邻

在所描述的场景中，您使用四个预测变量（收入、信用评分、就业和债务）来得出贷款批准的总体分数，您本质上是在使用线性回归模型。线性回归是一种监督机器学习算法，用于根据一个或多个输入特征预测连续输出变量。在本例中，您将根据四个预测变量的值来预测总体贷款审批分数。

第 4 题

You work for a political organization that does sentiment analysis of social media networks.Politicians look to your service to see how people feel about certain difficult topics. Yourorganization has developed an artificial neural network that can search social media fortopics and classify the comments as strongly agree, neutral, and strongly disagree. Whattype of machine learning are you using?You work in the marketing department for a largecompany, and you’d like to create a weekly opinion letter using ChatGPT to give your takeon the top news in your industry. You create a few test posts, and you notice that ChatGPTis getting the dates wrong and is mixing up the CEOs of different companies. Why are yourunning into this challenge?Your company wants to use generative Al to come up with newpharmaceuticals. This system will analyze all existing chemical compounds and try todevelop new compounds based on the success of some of your current pharmaceuticals.This system will require a lot of custom programming and access to your proprietary datasets. What type of generative Al system might work best?Your company wants to create asmartphone application that identifies plants using the phone’s camera. The companypurchases millions of digital images of plants labeled with the species names. You use thisinitial batch of images to train your artificial neural network, What type of machinelearning are you using for your network?

• supervised learning multiclass classification
• unsupervised learning binary classification
• variational auto encoding generative Al
• reinforcement learning unsupervised clustering

您在一家政治组织工作，该组织对社交媒体网络进行情绪分析。政治家希望通过您的服务来了解人们对某些困难话题的看法。您的组织开发了一种人工神经网络，可以在社交媒体中搜索主题并将评论分类为强烈同意、中立和强烈反对。您使用什么类型的机器学习？您在一家大公司的营销部门工作，您希望使用 ChatGpT 创建每周意见信，以表达您对行业头条新闻的看法。您创建了一些测试帖子，您注意到 ChatGpis 的日期错误，并且混淆了不同公司的首席执行官。你为什么要面对这个挑战？你的公司想要利用生成式人工智能来开发新的药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量自定义编程并访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 监督学习多类分类
• 无监督学习二元分类
• 变分自动编码生成算法
• 强化学习无监督聚类

所描述的场景涉及训练人工神经网络，将社交媒体的评论分为三类：强烈同意、中立和强烈不同意。这种类型的机器学习任务属于监督学习多类分类，其中向网络提供标记数据（带有相应情感标签的评论）来学习如何将新的、看不见的评论分类到预定义的类别中。

第 5 题

Your large social media company has decided to open source the data and source code foryour chatbot. You recently found out that a foreign government has downloaded your codeand set up a chatbot to spread propaganda. The chatbot encourages violence against anethnic minority group. What Al ethics violation might your chatbot release have caused?You work in the marketing department for a large company, and you’d like to create aweekly opinion letter using ChatGPT to give your take on the top news in your industry.You create a few test posts, and you notice that ChatGPT is getting the dates wrong and ismixing up the CEOs of different companies. Why are you running into this challenge?Yourcompany wants to use generative Al to come up with new pharmaceuticals. This system willanalyze all existing chemical compounds and try to develop new compounds based on thesuccess of some of your current pharmaceuticals, This system will require a lot of customprogramming and access to your proprietary data sets. What type of generative Al systemmight work best?Your company wants to create a smartphone application that identifiesplants using the phone’s camera. The company purchases millions of digital images ofplants labeled with the species names. You use this initial batch of images to train yourartificial neural network, What type of machine learning are you using for your network?

• Your technology assisted a human rights violation.
• There is now a danger of competition from a large well-funded government.
• Your technology is too easy to implement.
• Governments should always be able to use your technology for whatever reason they see fit.

您的大型社交媒体公司决定开源您的聊天机器人的数据和源代码。您最近发现外国政府下载了您的代码并设置了一个聊天机器人来传播宣传。该聊天机器人鼓励针对少数族裔群体的暴力行为。您的聊天机器人的发布可能会导致哪些违反道德的行为？您在一家大公司的营销部门工作，您希望使用 ChatGpT 创建每周意见信，以表达您对行业头条新闻的看法。您创建了一些意见书测试帖子，您会注意到 ChatGpT 的日期错误，并且混淆了不同公司的首席执行官。你为什么会遇到这个挑战？你的公司想要使用生成式人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前的一些药物的成功开发新的化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用第一批图像来训练您的人工神经网络，您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 你们的技术协助了侵犯人权的行为。
• 现在存在来自资金充足的大型政府的竞争的危险。
• 你们的技术太容易实现了。
• 政府应该始终能够出于他们认为合适的任何原因使用您的技术

所描述的情况是，聊天机器人被用来传播宣传并鼓励针对少数族裔群体的暴力行为，这构成了违反人工智能道德的行为，因为您的技术被用来促进侵犯人权。责任不仅在于创建和发布该技术，还在于考虑该技术可能如何被恶意行为者滥用。道德考虑对于防止此类有害后果至关重要。

第 6 题

You’re a director for an organization that detects credit card data. You’re trying to convinceyour manager to adopt a generative adversarial network (GAN) to test your system to see ifit can identify credit card fraud. What’s one of the best arguments you have for using thistype of neural network?You work in the marketing department for a large company, andyou’d like to create a weekly opinion letter using ChatGPT to give your take on the topnews in your industry. You create a few test posts, and you notice that ChatGpT is gettingthe dates wrong and is mixing up the CEOs of different companies. Why are you runninginto this challenge?Your company wants to use generative Al to come up with newpharmaceuticals. This system will analyze all existing chemical compounds and try todevelop new compounds based on the success of some of your current pharmaceuticals.This system will require a lot of custom programming and access to your proprietary datasets. What type of generative Al system might work best?Your company wants to create asmartphone application that identifies plants using the phone’s camera. The companypurchases millions of digital images of plants labeled with the species names. You use thisinitial batch of images to train your artificial neural network. What type of machinelearning are you using for your network?

• A GAN would allow the system to invent fraudulent transactions that aren’t present in the
• data.
• This type of system will generate many more fraudulent transactions than you would get witha typical neural network.
• Fraudulent transactions are by their very nature adversarial, so it’s good to have a networkthat reflects this.
• This type of neural network arrangement will be the easiest for your organization to set up.

您是一家检测信用卡数据的组织的主管。您正试图说服您的经理采用生成对抗网络 (GAN) 来测试您的系统，看看它是否可以识别信用卡欺诈。对于使用这种类型的神经网络，您的最佳论据是什么？您在一家大公司的营销部门工作，您希望使用 ChatGpT 创建每周意见信，以表达您对行业头条新闻的看法。您创建了一些测试帖子，然后您注意到 ChatGpT 的日期错误，并且混淆了不同公司的首席执行官。你为什么会遇到这个挑战？你的公司想要使用生成式人工智能来开发新的药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量自定义编程并访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• GAN 允许系统发明数据中不存在的欺诈交易。
• 与典型的神经网络相比，这种类型的系统会产生更多的欺诈交易。
• 欺诈交易本质上是对抗性的，因此拥有一个反映这一点的网络是件好事。
• 这种类型的神经网络安排对于您的组织来说是最容易设置的

使用生成对抗网络（GAN）检测信用卡欺诈的最佳论据之一是欺诈交易本质上是对抗性的。在欺诈检测的背景下，欺诈者正在积极尝试通过模仿合法交易来智胜和欺骗系统。 GAN 非常适合捕捉这种对抗性动态，因为它们涉及两个网络：一个试图创建虚假数据的生成器和一个试图区分真实数据和虚假数据的鉴别器。通过在真实交易和欺诈交易上训练 GAN，网络可以更好地学习与欺诈活动相关的复杂模式和行为，从而有可能提高欺诈检测能力。

第 7 题

Your company produces science fiction and fantasy graphic novels. One of your topillustrators has developed a style that is very strongly associated with your brand. Yourcompany decides to create a generative Al model to mimic their illustrations. This newmodel can create new graphics in their style in seconds. Now the company will have bettercontrol over their brand and increase productivity, What is one of the main challenges withthis approach?You work in the marketing department for a large company, and you’d like tocreate a weekly opinion letter using ChatGPT to give your take on the top news in yourindustry. You create a few test posts, and you notice that ChatGPT is getting the dateswrong and is mixing up the CEOs of different companies. Why are you running into thischallenge?Your company wants to use generative Al to come up with new pharmaceuticals.This system will analyze all existing chemical compounds and try to develop newcompounds based on the success of some of your current pharmaceuticals. This system willrequire a lot of custom programming and access to your proprietary data sets. What typeof generative Al system might work best?Your company wants to create a smartphoneapplication that identifies plants using the phone’s camera. The company purchasesmillions of digital images of plants labeled with the species names. You use this initial batchof images to train your artificial neural network, What type of machine learning are youusing for your network?

• Current generative Al models are not doing a very good job mimicking creative illustrators.
• The generative Al model will always need to be further trained, so it doesn’t save any time.
• It will “normalize mediocrity”_the graphics will look the same and lack a creative spark.
• lt is currently illegal in the United States to mimic the style of working illustrators.

贵公司制作科幻小说和奇幻图画小说。您的一位顶级插画师开发了一种与您的品牌密切相关的风格。你们的公司决定创建一个生成式人工智能模型来模仿他们的插图。这个新模型可以在几秒钟内创建出自己风格的新图形。现在，公司将更好地控制自己的品牌并提高生产力。这种方法的主要挑战之一是什么？您在一家大公司的营销部门工作，您希望使用 ChatGpT 创建每周意见信，以表达您对行业热门新闻的看法。您创建了一些测试帖子，然后您注意到 ChatGPT 的日期有误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要使用生成式人工智能来开发新的药物。该系统将分析所有现有的化合物，并尝试根据你当前的一些药物的成功开发新的化合物。该系统需要大量自定义编程和访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，可以使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 当前的生成式人工智能模型在模仿创意插画家方面表现不佳
• 生成式人工智能模型总是需要进一步训练，所以它不会节省任何时间。
• 这将使平庸“正常化”，图形看起来会一成不变，缺乏创意火花。
• 目前在美国模仿插画家的工作风格是违法的

使用生成式人工智能模型来模仿极具创造力的插画家的风格的主要挑战之一是它有可能“使平庸正常化”。这意味着，虽然模型可以有效地生成所需风格的图形，但生成的图形可能缺乏原始插画家为其作品带来的独特创意火花和创新。生成的内容可能会变得重复，并且无法捕捉人类艺术家贡献的相同水平的艺术表达和独特性。

第 8 题

What does the term model mean in generative Al?You work in the marketing departmentfor a large company, and you’d like to create a weekly opinion letter using ChatGPT to giveyour take on the top news in your industry. You create a few test posts, and you notice thatChatGPT is getting the dates wrong and is mixing up the CEOs of different companies. Whyare you running into this challenge?Your company wants to use generative Al to come upwith new pharmaceuticals. This system will analyze all existing chemical compounds andtry to develop new compounds based on the success of some of your currentpharmaceuticals. This system will require a lot of custom programming and access to yourproprietary data sets. What type of generative Al system might work best?Your companywants to create a smartphone application that identifies plants using the phone’s camera.The company purchases millions of digital images of plants labeled with the species names.You use this initial batch of images to train your artificial neural network. What type ofmachine learning are you using for your network?

• A model is Al mimicking human behavior.
• A model is a generative Al that trains another artificial intelligence on a dataset.
• A model is a set of algorithms that have been trained on a data set.
• A model is a data set of ethical issues.

术语“模型”在生成式 AI 中意味着什么？您在一家大公司的营销部门工作，您希望使用 ChatGpT 创建每周意见信，以表达您对行业热门新闻的看法。您创建了一些测试帖子，您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要使用生成式人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，可以使用手机摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的计算机人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 模型是人工智能模仿人类行为。
• 模型是一个生成式人工智能，它在数据集上训练另一个人工智能。
• 模型是一组在数据集上训练的算法
• 模型是道德问题的数据集。

在生成人工智能中，模型是指在数据集上训练的一组算法和参数。这种经过训练的模型能够生成类似于训练数据中存在的模式和特征的新内容，例如图像、文本或其他形式的数据。在生成式人工智能中训练模型的目标是捕获训练数据的基本分布，以便它可以生成与之前见过的相似的新颖且连贯的数据样本。

第 9 题

You work for a company that produces video games. One of the challenges is creating non.player characters (NPCs) that are controlled by the game, but still make strategic decisions.Your team decides to use machine learning, and each time an NpC does better than a playerit gets a small reward. Now the machine learning algorithms are coming up with interestingnew ways to play the game. What type of learning is this?You work in the marketingdepartment for a large company, and you’d like to create a weekly opinion letter usingChatGPT to give your take on the top news in your industry. You create a few test postsand you notice that ChatGpT is getting the dates wrong and is mixing up the CEOs ofdifferent companies. Why are you running into this challenge?Your company wants to usegenerative Al to come up with new pharmaceuticals. This system will analyze all existingchemical compounds and try to develop new compounds based on the success of some ofyour current pharmaceuticals. This system will require a lot of custom programming andaccess to your proprietary data sets. What type of generative Al system might work best?Your company wants to create a smartphone application that identifies plants using thephone’s camera. The company purchases millions of digital images of plants labeled withthe species names. You use this initial batch of images to train your artificial neuralnetwork. What type of machine learning are you using for your network?

• reinforcement learning
• generative Al
• self-supervised machine learning
• unsupervised machine learning

您在一家制作视频游戏的公司工作。挑战之一是创建由游戏控制但仍能做出战略决策的非玩家角色 (NPC)。您的团队决定使用机器学习，每当 NPC 比玩家做得更好时，就会获得少量奖励。现在，机器学习算法正在想出有趣的新游戏方式。这是什么类型的学习？您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周意见信，以表达您对行业热门新闻的看法。您创建了一些测试帖子。您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要利用再生人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前的一些药物的成功开发新的化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 强化学习
• 生成式人工智能
• 自监督机器学习
• 无监督机器学习

在强化学习中，代理（在本例中为 NPC）在环境中采取行动，以最大化累积奖励的某些概念。代理学习在不确定的、潜在复杂的环境中实现目标。

第 10 题

You are an executive of a company that is implementing generative Al systems. What arethe most essential ethical considerations to balance?You work in the marketing departmentfor a large company, and you’d like to create a weekly opinion letter using ChatGPT to giveyour take on the top news in your industry. You create a few test posts, and you notice thatChatGPT is getting the dates wrong and is mixing up the CEOs of different companies. Whyare you running into this challenge?Your company wants to use generative Al to come upwith new pharmaceuticals. This system will analyze all existing chemical compounds andtry to develop new compounds based on the success of some of your currentpharmaceuticals. This system will require a lot of custom programming and access to yourproprietary data sets. What type of generative Al system might work best?Your companywants to create a smartphone application that identifies plants using the phone’s camera.The company purchases millions of digital images of plants labeled with the species names.You use this initial batch of images to train your artificial neural network. What type ofmachine learning are you using for your network?

• your organization’s obligation to appease shareholders against your obligations to humanity
• the dangers to humanity against the possibility of your own enrichment
• getting creative generative Al output and optimizing production while maintaining humanoversight
• the cost of implementing these new systems against the costs of maintaining fullemployment

您是一家正在实施生成式人工智能系统的公司的高管。需要平衡的最重要的道德考虑因素是什么？您在一家大公司的营销部门工作，您希望使用 ChatGpT 创建每周意见书，以表达您对行业头条新闻的看法。您创建了一些测试帖子，您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要使用生成式人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，可以使用手机摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的计算机人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 您的组织有义务安抚股东，反对您对人性的义务
• 人类致富的可能性所面临的危险
• 获得创造性的生成性人工智能输出并优化生产，同时保持人工监督
• 实施这些新系统的成本与维持充分就业的成本

作为监督生成式人工智能系统实施的高管，需要平衡的最重要的道德考虑因素之一是获得创造性的生成式人工智能输出并优化生产，同时保持人类监督。虽然利用人工智能的能力来生成创意内容或优化生产流程很重要，但确保人工监督来审查和纠正人工智能生成的输出也同样重要。这有助于防止在未经任何检查的情况下产生潜在的偏见、错误或不道德的内容。

第 11 题

You are a software developer on a team that’s developing a generative Al nurse for ahealthcare company. You’ve trained the system on all your internal data, but to make itmore “worldly” you’ve also trained it with social media data. During your testing, youfound that sometimes the nurse will make recommendations that aren’t based on science.As a software developer, your Al ethical responsibility is to make sure that the Al nurse.You work in the marketing department for a large company, and you’d like to create aweekly opinion letter using ChatGPT to give your take on the top news in your industry.You create a few test posts, and you notice that ChatGpT is getting the dates wrong and ismixing up the CEOs of different companies. Why are you running into this challenge?Yourcompany wants to use generative Al to come up with new pharmaceuticals. This system willanalyze all existing chemical compounds and try to develop new compounds based on thesuccess of some of your current pharmaceuticals. This system will require a lot of customprogramming and access to your proprietary data sets. What type of generative Al systemmight work best?Your company wants to create a smartphone application that identifiesplants using the phone’s camera. The company purchases millions of digital images ofplants labeled with the species names. You use this initial batch of images to train yourartificial neural network, What type of machine learning are you using for your network?

• is always focused on generating data, increasing profits and reliable customer service
• is always using the latest information
• is developed in a way that’s transparent, explainable, and accountable
• is well-versed in alternative forms of treatment

您是一个团队的软件开发人员，该团队正在为一家医疗保健公司开发生成人工智能护士。您已经根据所有内部数据对系统进行了训练，但为了使其更加“世俗”，您还利用社交媒体数据对其进行了训练。在你的测试过程中，你发现有时护士会提出不基于科学的建议。作为一名软件开发人员，你的人工智能道德责任是确保人工智能护士。你在一家大公司的营销部门工作，并且您想使用 ChatGPT 创建每周意见书，以表达您对行业热门新闻的看法。您创建了一些测试帖子，并注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。你为什么会遇到这个挑战？你的公司想要使用生成式人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试在某些现有药物的成功基础上开发新化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 始终专注于生成数据、增加利润和可靠的客户服务
• 始终使用最新信息
• 以透明、可解释和负责任的方式开发
• 精通替代治疗形式

作为一名为医疗保健公司开发生成式人工智能护士的软件开发人员，您的道德责任是确保人工智能护士以透明、可解释和负责任的方式开发。这意味着人工智能系统应该为其建议提供清晰的推理，能够被医疗专业人员理解，并拥有适当的机制来跟踪决策的制定方式。这在医疗保健领域尤其重要，人工智能系统提出的建议可能对患者护理和结果产生重大影响。

第 12 题

What is a generative adversarial network (GAN)?You work in the marketing department fora large company, and you’d like to create a weekly opinion letter using ChatGPT to giveyour take on the top news in your industry. You create a few test posts, and you notice thatChatGPT is getting the dates wrong and is mixing up the CEOs of different companies. Whyare you running into this challenge?Your company wants to use generative Al to come upwith new pharmaceuticals. This system will analyze all existing chemical compounds andtry to develop new compounds based on the success of some of your currentpharmaceuticals. This system will require a lot of custom programming and access to yourproprietary data sets. What type of generative Al system might work best?Your companywants to create a smartphone application that identifies plants using the phone’s camera.The company purchases millions of digital images of plants labeled with the species names.You use this initial batch of images to train your artificial neural network. What type ofmachine learning are you using for your network?

• when two neural networks work cooperatively to produce the best output
• when a discriminator generates output so that a generator can review it and offer adversarialfeedback
• when two neural networks work in opposition, with a generator and a discriminator toimprove the generative output
• when two generative Al organizations compete for the same resources

什么是生成对抗网络 (GAN)？您在一家大公司的营销部门工作，您希望使用 ChatGpT 创建每周意见信，以表达您对行业热门新闻的看法。您创建了一些测试帖子，您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。您遇到了什么挑战？您的公司希望使用生成式人工智能来研制新药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，可以使用手机摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的计算机人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 当两个神经网络协同工作以产生最佳输出时
• 当鉴别器生成输出以便生成器可以对其进行审查并提供对抗反馈时
• 当两个神经网络反向工作时，使用生成器和判别器来提高生成输出
• 当两个生成人工智能组织争夺相同资源时

生成对抗网络（GAN）是一种机器学习模型，由两个神经网络（生成器和判别器）组成。这两个网络相互对抗以提高生成输出。生成器网络从随机噪声生成数据（例如图像、文本等），而鉴别器网络尝试区分训练数据集中的真实数据和生成器生成的假数据。通过这个竞争和反馈的过程，GAN 学会生成越来越真实和高质量的数据。

第 13 题

In machine learning, when a data model performs exceptionally well during the training setphase, but lacks the complexity to generate accurate predictions during the test set phase,the model is the data.You work in the marketing department for a large company, andyou’d like to create a weekly opinion letter using ChatGPT to give your take on the topnews in your industry, You create a few test posts, and you notice that ChatGPT is gettingthe dates wrong and is mixing up the CEOs of different companies. Why are you runninginto this challenge?Your company wants to use generative Al to come up with newpharmaceuticals. This system will analyze all existing chemical compounds and try todevelop new compounds based on the success of some of your current pharmaceuticals.This system will require a lot of custom programming and access to your proprietary datasets. What type of generative Al system might work best?Your company wants to create asmartphone application that identifies plants using the phone’s camera. The companypurchases millions of digital images of plants labeled with the species names. You use thisinitial batch of images to train your artificial neural network. What type of machinelearning are you using for your network?

• underfitting
• adversarial
• overfitting
• reinforcing

在机器学习中，当数据模型在训练集阶段表现得非常好，但在测试集阶段缺乏生成准确预测的复杂性时，该模型就是数据。您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周意见信，表达您对行业头条新闻的看法。您创建了一些测试帖子，然后您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。你为什么会遇到这个挑战？你的公司想要使用生成式人工智能来开发新的药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量自定义编程并访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器收益？

下面哪个选项是正确的

• 欠拟合
• 对抗
• 过拟合
• 强化

在机器学习中，当数据模型在训练集阶段表现异常良好，但在测试集阶段缺乏生成准确预测的复杂性时，该模型被称为欠拟合。当模型过于简单而无法捕获数据中的潜在模式时，就会出现欠拟合，从而导致模型无法很好地推广到新的和未见过的数据。

第 14 题

You recently purchased a new smartwatch. To set up the watch, you had to go to themanufacturer’s website and create a new account. When you create an account, it presentsa long license agreement you have to accept to create an account, You were anxious to useyour new watch, so you didn’t scroll through the 50-plus pages of the license agreement.What is one of the ethical issues with how the smartwatch manufacturer is operating?Youwork in the marketing department for a large company, and you’d like to create a weeklyopinion letter using ChatGPT to give your take on the top news in your industry. You createa few test posts, and you notice that ChatGPT is getting the dates wrong and is mixing upthe CEOs of different companies. Why are you running into this challenge?Your companywants to use generative Al to come up with new pharmaceuticals. This system will analyzeall existing chemical compounds and try to develop new compounds based on the successof some of your current pharmaceuticals. This system will require a lot of customprogramming and access to your proprietary data sets. What type of generative Al systemmight work best?Your company wants to create a smartphone application that identifiesplants using the phone’s camera. The company purchases millions of digital images ofplants labeled with the species names. You use this initial batch of images to train yourartificial neural network, What type of machine learning are you using for your network?

• It should have you automatically accept an agreement when you purchase the watch.
• lt should shorten the license agreement.
• lt should always try to keep the private data on the watch itself.
• They are not sharing a clear and transparent privacy policy.

您最近购买了一块新的智能手表。要设置手表，您必须访问制造商的网站并创建一个新帐户。当您创建帐户时，它会提供一份长许可协议，您必须接受该协议才能创建帐户。您急于使用您的新手表，因此没有滚动浏览 50 多页的许可协议。智能手表制造商的运营方式存在哪些道德问题？您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建一封每周意见信，以表达您对行业热门新闻的看法。您创建了一些测试帖子，然后您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。你为什么会遇到这个挑战？你的公司想要使用生成式人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前的一些药物的成功开发新的化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 当您购买手表时，它应该让您自动接受协议
• 它应该缩短许可协议。
• 它应该始终尝试将私人数据保留在手表本身上
• 他们没有共享清晰透明的隐私政策。

这家智能手表制造商的运营存在道德问题，因为他们提交了一份冗长的许可协议，但没有提供有关隐私政策和数据使用的清晰透明的信息。如果条款和条件不易于用户理解和访问，制造商可能无法充分告知用户他们的个人数据将如何被收集、存储和使用。这种缺乏透明度可能会导致用户在不知不觉中同意他们可能不完全理解的条款，这可能会对他们的隐私和数据安全产生影响。

第 15 题

What is the difference between generative Al and discriminative Al?You work in themarketing department for a large company, and you’d like to create a weekly opinion letterusing ChatGPT to give your take on the top news in your industry. You create a few testposts, and you notice that ChatGPT is getting the dates wrong and is mixing up the CEOs ofdifferent companies. Why are you running into this challenge?Your company wants to usegenerative Al to come up with new pharmaceuticals. This system will analyze all existingchemical compounds and try to develop new compounds based on the success of some ofyour current pharmaceuticals. This system will require a lot of custom programming andaccess to your proprietary data sets. What type of generative Al system might work best?Your company wants to create a smartphone application that identifies plants using thephone’s camera. The company purchases millions of digital images of plants labeled withthe species names. You use this initial batch of images to train your artificial neuralnetwork, What type of machine learning are you using for your network?

• Discriminative Al is mostly used in government and university work.
• Generative Al creates content while discriminative Al classifies data.
• Generative Al tends to not work with digital data.
• Discriminative Al creates content while generative Al classifies data

生成式人工智能和判别式人工智能有什么区别？您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周意见信，以表达您对行业热门新闻的看法。您创建了一些测试帖子，然后您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要利用再生人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前的一些药物的成功开发新的化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 歧视性人工智能主要用于政府和大学工作。
• 生成式人工智能创建内容，而判别式人工智能对数据进行分类
• 生成式人工智能往往不适用于数字数据。
• 判别式人工智能创建内容，而生成式人工智能对数据进行分类

生成式人工智能和判别式人工智能是两种不同类型的机器学习方法：

• 生成式人工智能：这种类型的人工智能旨在创建与现有示例类似的新数据，例如图像、文本或音频。它对数据的基本分布进行建模，并根据该分布生成新样本。例如，生成对抗网络 (GAN) 和变分自动编码器 (VAE) 等生成模型用于创建类似于训练数据的真实图像或文本。
• 判别式人工智能：此类人工智能专注于分类和决策任务。判别模型旨在学习分隔不同类别数据的边界或决策边界。他们专注于将输入数据映射到输出标签或类别。判别模型的常见示例包括逻辑回归、支持向量机以及用于图像分类或文本情感分析的深度神经网络。

第 16 题

You are a technical manager for a large city courthouse. The judges have asked you toimplement a new system that will make criminal sentencing recommendations. As part ofyour testing, your team has the system make sentencing recommendations for past courtconvictions. Your team finds that the new system is much more likely to recommend longesentences for some groups of people. What is the main ethical challenge withimplementing this system?You work in the marketing department for a large company, aneyou’d like to create a weekly opinion letter using ChatGPT to give your take on the topnews in your industry. You create a few test posts, and you notice that ChatGPT is gettingthe dates wrong and is mixing up the CEOs of different companies. Why are you runninginto this challenge?Your company wants to use generative Al to come up with newpharmaceuticals. This system will analyze all existing chemical compounds and try todevelop new compounds based on the success of some of your current pharmaceuticals.This system will require a lot of custom programming and access to your proprietary datasets. What type of generative Al system might work best?Your company wants to create asmartphone application that identifies plants using the phone’s camera. The companypurchases millions of digital images of plants labeled with the species names. You use thisinitial batch of images to train your artificial neural network, What type of machinelearning are you using for your network?

• It magnifies existing biases rather than mitigating them.
• The courthouse obviously does not have the technical expertise to improve the system.
• The city courthouse might not be able to afford the service.
• Impartial judges should make sentencing recommendations. Al systems should not beinvolved.

您是一家大城市法院的技术经理。法官要求您实施一个新系统，以提出刑事量刑建议。作为测试的一部分，您的团队让系统对过去的法庭定罪提出量刑建议。您的团队发现新系统更有可能为某些人群推荐长句子。实施该系统的主要道德挑战是什么？您在一家大公司的营销部门工作，您希望使用 ChatGpT 创建每周意见信，以表达您对行业头条新闻的看法。您创建了一些测试帖子，然后您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。你为什么会遇到这个挑战？你的公司想要使用生成式人工智能来开发新的药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量自定义编程并访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器收益？

下面哪个选项是正确的

• 它放大了现有的偏见，而不是减轻它们。
• 法院显然不具备改进系统的技术专长
• 市法院可能无法承担这项服务。
• 法官应公正地提出量刑建议。人工智能系统不应该参与其中。

实施建议对某些人群判处更长刑期的新制度的主要道德挑战是，它可能会放大刑事司法系统中现有的偏见。如果系统接受了反映种族或社会经济差异等偏见的历史数据的训练，那么人工智能系统就可以学习并在其建议中延续这些偏见。这可能导致不公平和不公正的结果，加剧司法系统中现有的不平等。

第 17 题

Why did early artificial intelligence systems do so well with board games?You work in themarketing department for a large company, and you’d like to create a weekly opinion letterusing ChatGPT to give your take on the top news in your industry. You create a few testposts, and you notice that ChatGPT is getting the dates wrong and is mixing up the CEOs ofdifferent companies. Why are you running into this challenge?Your company wants to usegenerative Al to come up with new pharmaceuticals. This system will analyze all existingchemical compounds and try to develop new compounds based on the success of some ofyour current pharmaceuticals. This system will require a lot of custom programming andaccess to your proprietary data sets. What type of generative Al system might work best?Your company wants to create a smartphone application that identifies plants using thephone’s camera. The company purchases millions of digital images of plants labeled withthe species names. You use this initial batch of images to train your artificial neuralnetwork. What type of machine learning are you using for your network?

• Because even with their limiting processing power, early systems thrived in a world of simplerules and pattern matching.
• Because board games give the system unique insight into human behavior, early systemscould learn and mimic the same behavior.
• Because computer scientists could do a good job programming all the rules into the gamethat the system would understand.
• Because board games are inherently chaotic, the system had a lot of opportunities to crunchnew data.

为什么早期的人工智能系统在棋盘游戏方面表现得如此出色？您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周意见信，以表达您对行业头条新闻的看法。您创建了一些测试帖子，然后您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要利用再生人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前的一些药物的成功开发新的化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 因为即使处理能力有限，早期系统仍然在简单规则和模式匹配的世界中蓬勃发展。
• 由于棋盘游戏为系统提供了对人类行为的独特洞察，因此早期系统可以学习并模仿相同的行为。
• 因为计算机科学家可以很好地将所有规则编写到系统能够理解的游戏中。
• 由于棋盘游戏本质上是混乱的，因此系统有很多机会处理新数据。

早期的人工智能系统在棋盘游戏中表现良好，因为这些游戏通常具有明确的规则和有限的复杂性。这使得处理能力和能力有限的早期人工智能系统可以通过使用基于规则的方法和模式匹配来脱颖而出。棋盘游戏的确定性使人工智能系统能够探索所有可能的动作和结果，从而促进战略决策。

第 18 题

You just purchased a new smart music player that can add songs if you simply describe thecharacteristics you’re looking for, To test out the player you say, “Will you add that famousJohnny Cash song that they turned into a movie?” What type of technology will your musicplayer use to find your song?You work in the marketing department for a large company,and you’d like to create a weekly opinion letter using ChatGPT to give your take on the topnews in your industry. You create a few test posts, and you notice that ChatGPT is gettingthe dates wrong and is mixing up the CEOs of different companies. Why are you runninginto this challenge?Your company wants to use generative Al to come up with newpharmaceuticals. This system will analyze all existing chemical compounds and try todevelop new compounds based on the success of some of your current pharmaceuticals.This system will require a lot of custom programming and access to your proprietary datasets. What type of generative Al system might work best?Your company wants to create asmartphone application that identifies plants using the phone’s camera. The companypurchases millions of digital images of plants labeled with the species names. You use thisinitial batch of images to train your artificial neural network. What type of machinelearning are you using for your network?

• It is using unsupervised machine learning to make a recommendation based on your musictastes.
• lt is using reinforcement learning to create a personalized playlist for your music player.
• It is converting your description into a search by using natural language processing.
• It is using generative Al to compose a new song.

您刚刚购买了一个新的智能音乐播放器，如果您简单地描述您正在寻找的特征，它就可以添加歌曲。为了测试播放器，你可以说：“你会添加他们改编成电影的那首著名的约翰尼·卡什歌曲吗？”您的音乐播放器将使用什么类型的技术来查找您的歌曲？您在一家大公司的营销部门工作。您希望使用 ChatGpT 创建每周意见信，以表达您对行业头条新闻的看法。您创建了一些测试帖子，然后您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。你为什么会遇到这个挑战？你的公司想要使用生成式人工智能来开发新的药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量自定义编程并访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 它使用无监督机器学习根据您的音乐品味提出推荐。
• 它正在使用强化学习为您的音乐播放器创建个性化播放列表。
• 它正在使用自然语言处理将您的描述转换为搜索。
• 它正在使用生成式 Al 来创作一首新歌

让我们分解一下问题中提出的场景：

• 智能音乐播放器：音乐播放器试图理解用户的描述来查找特定的歌曲。这涉及理解人类语言并将其转换为有意义的查询来搜索歌曲。这里使用的技术是自然语言处理（NLP）。
• ChatGPT 的错误：ChatGPT 与任何其他模型一样，都是基于其训练数据的。如果日期错误或混淆了首席执行官，很可能是因为模型的训练数据可能不是最新的，或者它可能根据所看到的大量信息犯了错误。这是模型训练及其所接触的数据量巨大的限制。
• 制药的生成人工智能：对于基于现有化合物创建新的药物化合物，生成模型将是理想的选择。具体来说，生成对抗网络（GAN）在生成与训练数据相似但不完全相同的新数据方面表现出了希望。在这种情况下，GAN 可能会根据数据集中成功的化合物生成新的化合物。
• 用于植物识别的智能手机应用程序：当您拥有大型标记数据集（在本例中为带有物种名称的植物图像）并使用它来训练模型时，您正在使用监督机器学习。该模型学习将图像的特定特征与相应的标签（物种名称）相关联。

第 19 题

You’re an executive for a software development company. Your company develops only oneproduct. You want to include ethical decision-making into your software development, soyou ask a senior developer to also serve as the company’s chief Al ethics officer. Whatwould be one of the challenges with this approach?You work in the marketing departmentfor a large company, and you’d like to create a weekly opinion letter using ChatGPT to giveyour take on the top news in your industry. You create a few test posts, and you notice thatChatGPT is getting the dates wrong and is mixing up the CEOs of different companies. Whyare you running into this challenge?Your company wants to use generative Al to come upwith new pharmaceuticals. This system will analyze all existing chemical compounds andtry to develop new compounds based on the success of some of your currentpharmaceuticals. This system will require a lot of custom programming and access to yourproprietary data sets. What type of generative Al system might work best?Your companywants to create a smartphone application that identifies plants using the phone’s camera.The company purchases millions of digital images of plants labeled with the species names.You use this initial batch of images to train your artificial neural network. What type ofmachine learning are you using for your network?

• Since you have only one product, your senior developer should always be focusing onsoftware development.
• With only one product, there aren’t going to be many ethical Al issues, so you should haveyour developers focus on developing software.
• A chief Al ethics officer sets the ethical direction for the entire company and shouldn’t justfocus on the product.
• Software developers are busy people and they need to focus on technical challenges.

您是一家软件开发公司的高管。你们公司只开发一种产品。您希望将道德决策纳入您的软件开发中，因此您要求一位高级开发人员兼任公司的首席人工智能道德官。这种方法面临的挑战之一是什么？您在一家大公司的营销部门工作，您希望使用 ChatGpPT 创建每周意见信，以表达您对行业热门新闻的看法。您创建了一些测试帖子，您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要使用生成式人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 由于您只有一种产品，因此您的高级开发人员应该始终专注于软件开发。
• 只有一种产品，不会出现太多道德人工智能问题，因此您应该让开发人员专注于开发软件。
• 首席人工智能道德官为整个公司设定道德方向，不应只关注产品。
• 软件开发人员很忙，他们需要关注技术挑战

第 20 题

You work on a team that’s developing a generative Al text-to-image service. Your servicewill specialize in creating realistic looking paintings. To train the system you have toprocess millions of digitized paintings. The system has learned that paintings almost alwayshave a signature. When you test the system, it creates a fake signature on the painting. Theproduct manager asks you to create an algorithm to remove the signatures. What might bean ethical challenge with this approach?You work in the marketing department for a largecompany, and you’d like to create a weekly opinion letter using ChatGPT to give your takeon the top news in your industry. You create a few test posts, and you notice that ChatGPTis getting the dates wrong and is mixing up the CEOs of different companies. why are yourunning into this challenge?Your company wants to use generative Al to come up with newpharmaceuticals. This system will analyze all existing chemical compounds and try todevelop new compounds based on the success of some of your current pharmaceuticals.This system will require a lot of custom programming and access to your proprietary datasets. What type of generative Al system might work best?Your company wants to create asmartphone application that identifies plants using the phone’s camera. The companypurchases millions of digital images of plants labeled with the species names. You use thisinitial batch of images to train your artificial neural network. What type of machinelearning are you using for your network?

• If it signs the painting, then it might have questionable intellectual property rights.
• It’s a missed opportunity to show the artistic strength of generative Al.
• It isn’t transparent in how the system collects its data.
• lt should allow the system to create a fake form of attribution.

您所在的团队正在开发生成式人工智能文本到图像服务。您的服务将专注于创作逼真的绘画。要训练该系统，您必须处理数百万幅数字化绘画。系统了解到，画作几乎总是有签名。当您测试系统时，它会在画作上创建一个假签名。产品经理要求您创建一个算法来删除签名。这种方法可能会带来什么道德挑战？您在一家大公司的营销部门工作，您希望使用 ChatGpT 创建每周意见书，以表达您对行业头条新闻的看法。您创建了一些测试帖子，您注意到 ChatGpis 的日期错误，并且混淆了不同公司的首席执行官。你为什么要面对这个挑战？你的公司想要利用生成式人工智能来开发新的药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量自定义编程并访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 如果它在这幅画上签名，那么它的知识产权可能有问题。
• 这是一个展示生成性阿尔的艺术实力的机会。
• 系统收集数据的方式并不透明。
• 它应该允许系统创建虚假的归因形式。

第 21 题

When you’ re approached with a generative Al ethics challenge, what is one of the firstquestions you should ask?You work in the marketing department for a large company, andyou’d like to create a weekly opinion letter using ChatGPT to give your take on the topnews in your industry. You create a few test posts, and you notice that ChatGPT is gettingthe dates wrong and is mixing up the CEOs of different companies. Why are you runninginto this challenge?Your company wants to use generative Al to come up with newpharmaceuticals. This system will analyze all existing chemical compounds and try todevelop new compounds based on the success of some of your current pharmaceuticals.This system will require a lot of custom programming and access to your proprietary datasets. What type of generative Al system might work best?Your company wants to create asmartphone application that identifies plants using the phone’s camera. The companypurchases millions of digital images of plants labeled with the species names. You use thisinitial batch of images to train your artificial neural network. What type of machinelearning are you using for your network?

• How can you get this product to market quickly?
• What would cause the least harm to the greatest number of people?
• What would be most profitable for your organization?
• What is the highest standard of responsible human behavior?

当您面临生成式人工智能道德挑战时，您应该问的第一个问题是什么？您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周一封意见书来表达您的看法您所在行业的头条新闻。您创建了一些测试帖子，然后您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。你为什么会遇到这个挑战？你的公司想要使用生成式人工智能来开发新的药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量自定义编程并访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 如何才能将该产品快速推向市场？
• 什么会对最多的人造成最小的伤害？
• 什么对您的组织来说最有利可图？
• 负责任的人类行为的最高标准是什么？

第 22 题

You use a new text-to-image generating service to create a beautiful artistic landscape. Yolsubmit your artwork for an Al-generated artistic award and win third place. The newsmedia picks up the story and uses your image in online articles without any compensationor attribution. Did they violate your copyright protection?You work in the marketingdepartment for a large company, and you’d like to create a weekly opinion letter usingChatGPT to give your take on the top news in your industry. You create a few test posts.and you notice that ChatGPT is getting the dates wrong and is mixing up the CEOs ofdifferent companies. Why are you running into this challenge?Your company wants to usegenerative Al to come up with new pharmaceuticals. This system will analyze all existingchemical compounds and try to develop new compounds based on the success of some ofyour current pharmaceuticals. This system will require a lot of custom programming andaccess to your proprietary data sets. What type of generative Al system might work best?Your company wants to create a smartphone application that identifies plants using thephone’s camera. The company purchases millions of digital images of plants labeled withthe species names. You use this initial batch of images to train your artificial neuralnetwork. What type of machine learning are you using for your network?

• No, they didn’t violate your copyright protection, but they did violate the system’s.
• Yes, your image and the prompt engineering phrase can be protected by copyright.
• No, because currently Al-generated images can’t be protected by copyright.
• Yes, but you’ll have to split the copyright proceeds and attribution with the Al system.

您可以使用新的文本到图像生成服务来创建美丽的艺术景观。您将您的作品提交给 Al 生成的艺术奖并获得第三名。新闻媒体会报道这个故事，并在在线文章中使用您的图像，而无需任何补偿或归属。他们是否侵犯了您的版权保护？您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周意见信，以表达您对行业头条新闻的看法。您创建了一些测试帖子。您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要利用再生人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前的一些药物的成功开发新的化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 不，他们没有侵犯你的版权保护，但他们确实侵犯了系统的版权保护。
• 是的，您的图像和提示工程短语可以受版权保护。
• 不可以，因为目前 AI 生成的图像无法受到版权保护。
• 是的，但你必须与 Al 系统分割版权收益和归属。

第 23 题

You’re trying to improve your skills with prompt engineering, so you asked ChatGPT togenerate a paragraph of text. The first prompt you create is, “Tell me about lactoseintolerance.” You weren’t satisfied with the results, so for the second prompt you wrote,”Write a blog article on lactose intolerance for my healthcare website.” What did you dowith the second prompt that you didn’t do with the first?You work in the marketingdepartment for a large company, and you’d like to create a weekly opinion letter usingChatGPT to give your take on the top news in your industry. You create a few test posts.and you notice that ChatGPT is getting the dates wrong and is mixing up the CEOs ofdifferent companies. Why are you running into this challenge?Your company wants to usegenerative Al to come up with new pharmaceuticals. This system will analyze all existingchemical compounds and try to develop new compounds based on the success of some ofyour current pharmaceuticals. This system will require a lot of custom programming andaccess to your proprietary data sets. What type of generative Al system might work best?Your company wants to create a smartphone application that identifies plants using thephone’s camera. The company purchases millions of digital images of plants labeled withthe species names. You use this initial batch of images to train your artificial neuralnetwork. What type of machine learning are you using for your network?

• You asked for an adversarial response.
• You provided context.
• You used an analogy.
• You started a brainstorming session.

您试图通过提示工程来提高您的技能，因此您要求 ChatGPT 生成一段文本，您创建的第一个提示是“告诉我有关乳糖不耐受的信息”。您对结果不满意，因此对于第二个提示，您写道“为我的医疗保健网站写一篇有关乳糖不耐受的博客文章”。您对第二个提示做了什么，而对第一个提示没有做？您在一家大公司的营销部门工作，您想使用 ChatGPT 创建每周意见信，以表达您对行业热门新闻的看法。您创建了一些测试帖子，您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要利用再生人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前的一些药物的成功开发新的化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 你要求做出对抗性的回应。
• 您提供了背景。
• 你用了一个比喻。
• 您开始了头脑风暴会议

第 24 题

How is an artificial neural network related to machine learning?You work in the marketingdepartment for a large company, and you’d like to create a weekly opinion letter usingChatGPT to give your take on the top news in your industry. You create a few test postsand you notice that ChatGPT is getting the dates wrong and is mixing up the CEOs ofdifferent companies. Why are you running into this challenge?Your company wants to usegenerative Al to come up with new pharmaceuticals. This system will analyze all existingchemical compounds and try to develop new compounds based on the success of some ofyour current pharmaceuticals. This system will require a lot of custom programming andaccess to your proprietary data sets. What type of generative Al system might work best?Your company wants to create a smartphone application that identifies plants using thephone’s camera. The company purchases millions of digital images of plants labeled withthe species names. You use this initial batch of images to train your artificial neuralnetwork. What type of machine learning are you using for your network?

• An artificial neural network is a machine learning technique.
• An artificial neural network uses preprogrammed responses instead of learning
• An artificial neural network does not require programming like a machine learning system.
• An artificial neural network is an earlier form of machine learning.

人工神经网络与机器学习有何关系？您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周意见信，以表达您对行业热门新闻的看法。您创建了一些测试帖子，您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要利用再生人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前的一些药物的成功开发新的化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 人工神经网络是一种机器学习技术。
• 人工神经网络使用预编程的响应而不是学习。
• 人工神经网络不需要像机器学习系统那样进行编程。
• 人工神经网络是机器学习的早期形式。

第 25 题

You work for a large financial institution that wants to identify undervalued stocks. To doso, you feed decades of financial information into an artificial neural network to createclusters of stocks. Then your data science team tries to find stocks in those clusters thatsubstantially increased in value. Your data science team hopes to find stocks in the samecluster that may also gain value. What type of machine learning are you using?You work inthe marketing department for a large company, and you’d like to create a weekly opinionletter using ChatGPT to give your take on the top news in your industry. You create a fewtest posts, and you notice that ChatGPT is getting the dates wrong and is mixing up theCEOs of different companies. Why are you running into this challenge?Your company wantsto use generative Al to come up with new pharmaceuticals. This system will analyze allexisting chemical compounds and try to develop new compounds based on the success ofsome of your current pharmaceuticals. This system will require a lot of customprogramming and access to your proprietary data sets. What type of generative Al systemmight work best?Your company wants to create a smartphone application that identifiesplants using the phone’s camera. The company purchases millions of digital images ofplants labeled with the species names. You use this initial batch of images to train yourartificial neural network, What type of machine learning are you using for your network?

• reinforcement learning
• unsupervised learning
• supervised machine learning
• generative artificial intelligence

您在一家大型金融机构工作，想要识别被低估的股票。为此，您需要将数十年的金融信息输入人工神经网络以创建股票集群。然后，您的数据科学团队尝试在这些集群中寻找价值大幅增加的股票。您的数据科学团队希望在同一集群中找到也可能增值的股票。您使用什么类型的机器学习？您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周评论员，以表达您对行业热门新闻的看法。您创建了一些测试帖子，然后您注意到 ChatGpT 的日期错误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要使用生成式人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 强化收益
• 无监督学习
• 监督机器学习
• 生成式人工智能

第 26 题

You are having some difficulty dealing with a colleague at work. You asked ChatGPT foradvice on how to improve your relationship with the coworker. ChatGPT gives youextremely helpful advice. You find yourself intuitively thanking it for its help. Given yourinteraction, do you think that ChatGPT is strong or weak Al?You work in the marketingdepartment for a large company, and you’d like to create a weekly opinion letter usingChatGPT to give your take on the top news in your industry. You create a few test posts,and you notice that ChatGPT is getting the dates wrong and is mixing up the CEOs ofdifferent companies. Why are you running into this challenge?Your company wants to usegenerative Al to come up with new pharmaceuticals. This system will analyze all existingchemical compounds and try to develop new compounds based on the success of some ofyour current pharmaceuticals. This system will require a lot of custom programming andaccess to your proprietary data sets. What type of generative Al system might work best?Your company wants to create a smartphone application that identifies plants using thephone’s camera. The company purchases millions of digital images of plants labeled withthe species names. You use this initial batch of images to train your artificial neuralnetwork. What type of machine learning are you using for your network?

• lt is weak Al because ChatGPT is a good example of artificial general intelligence.
• It is weak Al because ChatGPT doesn’t understand what it’s saying it’s just gatheringinformation that it found online.
• It is strong Al because ChatGPT gave you genuinely helpful advice that’s the same quality as ahuman s.
• It is neither because ChatGPT is a generative Al system which falls outside the distinctions intraditional artificial intelligence.

您在工作中与同事相处时遇到一些困难。您向 ChatGPT 寻求有关如何改善与同事关系的建议。 ChatGPT 为您提供非常有用的建议。你会发现自己直觉地感谢它的帮助。考虑到您的互动，您认为 ChatGPT 强还是弱？您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周意见书，以表达您对行业热门新闻的看法。您创建了一些测试帖子。您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要利用再生人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前的一些药物的成功开发新的化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 它是弱人工智能，因为 ChatGPT 是通用人工智能的一个很好的例子。
• 它是弱人工智能，因为 ChatGPT 不明白它在说什么，它只是收集在网上找到的信息。
• 它很强大，因为 ChatGPT 为您提供了真正有用的建议，其质量与人类的建议相同
• 这也不是因为 ChatGPT 是一个生成式人工智能系统，不属于传统人工智能的区别。

第 27 题

Your company wants to use generative Al to come up with new pharmaceuticals. Thissystem will analyze all existing chemical compounds and try to develop new compoundsbased on the success of some of your current pharmaceuticals. This system will require a lotof custom programming and access to your proprietary data sets. What type of generativeAl system might work best?Your company wants to create a smartphone application thatidentifies plants using the phone’s camera. The company purchases millions of digitalimages of plants labeled with the species names. You use this initial batch of images totrain your artificial neural network, What type of machine learning are you using for yournetwork?

• Combine a series of open-source models and run on a cloud service.
• Develop your own generative Al model based on your existing data.
• Use a generative Al service like ChatGPT.
• Use a text to graphics engine such as DALL-E 2.

贵公司希望利用生成式人工智能来研制新药物。该系统将分析所有现有的化合物，并尝试根据您当前的一些药物的成功开发新的化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 结合一系列开源模型并运行在coud服务上
• 根据现有数据开发您自己的生成式人工智能模型
• 使用像 ChatGPT 这样的生成式 AI 服务
• 使用文本到图形引擎，例如 DALL-E 2

第 28 题

You work in the marketing department for a large company, and you’d like to create aweekly opinion letter using ChatGPT to give your take on the top news in your industry.You create a few test posts, and you notice that ChatGPT is getting the dates wrong and ismixing up the CEOs of different companies. Why are you running into this challenge?Yourcompany wants to use generative Al to come up with new pharmaceuticals. This system willanalyze all existing chemical compounds and try to develop new compounds based on thesuccess of some of your current pharmaceuticals. This system will require a lot of customprogramming and access to your proprietary data sets. What type of generative Al systemmight work best?Your company wants to create a smartphone application that identifiesplants using the phone’s camera. The company purchases millions of digital images ofplants labeled with the species names. You use this initial batch of images to train yourartificial neural network. What type of machine learning are you using for your network?

• ChatGPT shouldn’t be used for creative writing because it’s still prone to factual errors.
• ChatGPT is getting much better at opinion-based writing, so you should use it now to getahead of the game.
• ChatGPT needs to scale up so that it has a better understanding of your industry.
• There’s a good chance that these are human errors that can be corrected by fully embracingChatGPT.

您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周意见书，以表达您对行业热门新闻的看法。您创建了一些测试帖子，并且注意到 ChatGpT 正在获得日期错误，并且混淆了不同公司的首席执行官。你为什么会遇到这个挑战？你的公司想要使用生成式人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前的一些药物的成功开发新的化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• ChatGPT 不应用于创意写作，因为它仍然容易出现事实错误
• ChatGPT 在基于意见的写作方面做得越来越好，因此您现在应该使用它来取得领先地位。
• ChatGPT 需要扩大规模，以便更好地了解您的行业
• 这些很有可能是人为错误，可以通过完全 embracinoChatGPT 来纠正

第 29 题

You work for a large credit card company that wants to create an artificial neural networkthat will help predict when people are going to have trouble paying their bills. so yourteam gathers all the billing statements for people who had trouble paying their bills. Thenyou feed this data into an artificial neural network, What is this process called?You work inthe marketing department for a large company, and you’d like to create a weekly opinionletter using ChatGPT to give your take on the top news in your industry. You create a fewtest posts, and you notice that ChatGPT is getting the dates wrong and is mixing up theCEOs of different companies. Why are you running into this challenge?Your company wantsto use generative Al to come up with new pharmaceuticals. This system will analyze allexisting chemical compounds and try to develop new compounds based on the success ofsome of your current pharmaceuticals. This system will require a lot of customprogramming and access to your proprietary data sets. What type of generative Al systemmight work best?Your company wants to create a smartphone application that identifiesplants using the phone’s camera. The company purchases millions of digital images ofplants labeled with the species names. You use this initial batch of images to train yourartificial neural network. What type of machine learning are you using for your network?

• This is testing your artificial neural network with unlabeled data.
• This is classifying your data using reinforcement labels.
• This is training your artificial neural network with labeled data.
• This is unsupervised machine learning.

您在一家大型信用卡公司工作，该公司希望创建一个人工神经网络，帮助预测人们何时会遇到支付账单的困难。因此，您的团队会收集无法支付账单的人们的所有账单。然后将这些数据输入人工神经网络。这个过程叫什么？您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周意见信，以表达您对行业头条新闻的看法。您创建了一些测试帖子，然后注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要使用生成式人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 这是使用未标记的数据测试您的人工神经网络
• 这是使用强化标签对数据进行分类。
• 这是使用标记数据训练您的人工神经网络。
• 这是无监督机器学习

第 30 题

You are an executive for a large company that has a customer service department. Recentlysome of the top managers have been talking about replacing customer servicerepresentatives with an Al chatbot. Some of the managers feel like the chatbot shouldimpersonate a human customer service representative. They argue that if customers knowit’s an Al chatbot then they would immediately disconnect, Other managers feel like itwould be unethical to impersonate a human. What would be the best place tocommunicate your decision?You work in the marketing department for a large company,and you’d like to create a weekly opinion letter using ChatGPT to give your take on the topnews in your industry. You create a few test posts, and you notice that ChatGPT is gettingthe dates wrong and is mixing up the CEOs of different companies. Why are you runninginto this challenge?Your company wants to use generative Al to come up with newpharmaceuticals. This system will analyze all existing chemical compounds and try todevelop new compounds based on the success of some of your current pharmaceuticals.This system will require a lot of custom programming and access to your proprietary datasets. What type of generative Al system might work best?Your company wants to create asmartphone application that identifies plants using the phone’s camera. The companypurchases millions of digital images of plants labeled with the species names. You use thisinitial batch of images to train your artificial neural network. What type of machinelearning are you using for your network?

• Create a Responsible Al Policy and Governance framework.
• The executives should leave it to the product development team.
• Schedule a company-wide meeting.
• Create a Profitability with Generative Al Action Plan document.

您是一家设有客户服务部门的大公司的高管。最近，一些高层管理人员一直在讨论用 A 聊天机器人取代客户服务代表。一些经理认为聊天机器人应该模仿人类客户服务代表。他们认为，如果客户知道这是一个 Al 聊天机器人，那么他们会立即断开连接。其他经理认为冒充人类是不道德的。传达您的决定的最佳地点是什么？您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周意见书，以表达您对行业头条新闻的看法。您创建了一些测试帖子，然后您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。你为什么会遇到这个挑战？你的公司想要使用生成式人工智能来开发新的药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量自定义编程并访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 创建负责任的人工智能政策和治理框架。
• 高管们应该把它留给产品开发团队。
• 安排全公司会议
• 使用生成式 AI 行动计划文件创建盈利能力。

该框架将为公司内部人工智能的道德使用提供明确的指导方针，确保人工智能部署的透明度、问责制和可信度。它还可以作为未来人工智能相关决策和项目的参考。

第 31 题

Your online movie-streaming business wants to create an artificial neural network that canrecommend new movies based on what customers have already seen. The team creates aseries of xY diagrams of different film genres. Then it puts the film rating along the X-axisand the duration that people watch on the Y-axis. it then makes a recommendation basedon how close movies are to each other on the chart. What type of machine learningalgorithm is the team using?You work in the marketing department for a large company.and you’d like to create a weekly opinion letter using ChatGPT to give your take on the topnews in your industry. You create a few test posts, and you notice that ChatGPT is gettingthe dates wrong and is mixing up the CEOs of different companies. Why are you runninginto this challenge?Your company wants to use generative Al to come up with newpharmaceuticals. This system will analyze all existing chemical compounds and try todevelop new compounds based on the success of some of your current pharmaceuticals.This system will require a lot of custom programming and access to your proprietary datasets. What type of generative Al system might work best?Your company wants to create asmartphone application that identifies plants using the phone’s camera. The companypurchases millions of digital images of plants labeled with the species names. You use thisinitial batch of images to train your artificial neural network. What type of machinelearning are you using for your network?

• K-nearest neighbor
• reinforcement learning
• Q leamming
• Naive Bayes

您的在线电影流媒体业务想要创建一个人工神经网络，根据客户已经看过的内容推荐新电影。该团队创建了一系列不同电影类型的 XY 图。然后，它将电影评级放在 X 轴上，将人们观看的时长放在 Y 轴上。然后，它会根据图表上电影之间的接近程度做出推荐。团队使用什么类型的机器学习算法？您在一家大公司的营销部门工作。您希望使用 ChatGpT 创建每周意见信，以表达您对行业头条新闻的看法。您创建了一些测试帖子，然后您注意到 ChatGpT 的日期错误，并且混淆了不同公司的首席执行官。你为什么会遇到这个挑战？你的公司想要使用生成式人工智能来开发新的药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量自定义编程并访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• K-最近邻
• 强化学习
• 问学习
• 朴素贝叶斯

K 最近邻 (K-NN) 算法是一种基于实例的学习，其中函数在局部近似，并且所有计算都推迟到分类。它用于特征空间中彼此靠近的数据点可能共享标签的情况。在所描述的电影推荐系统的背景下，XY图上彼此接近的电影（基于评级和观看持续时间）被认为是相似的并且可以被推荐给用户。

第 32 题

Oxford dictionary defines plagiarism as “the practice of taking someone else’s work orideas and passing them off as one’s own.” lf you ask ChatGPT to describe a sunset, it willgive you a response, but these systems have never experienced a sunset. The only way itcould respond is by “passing off ideas as its own.” Does that mean that these generative Alsystems are plagiarism machines?You work in the marketing department for a largecompany, and you’d like to create a weekly opinion letter using ChatGPT to give your takeon the top news in your industry. You create a few test posts, and you notice that ChatGPTis getting the dates wrong and is mixing up the CEOs of different companies. Why are yourunning into this challenge?Your company wants to use generative Al to come up with newpharmaceuticals. This system will analyze all existing chemical compounds and try todevelop new compounds based on the success of some of your current pharmaceuticals.This system will require a lot of custom programming and access to your proprietary datasets. What type of generative Al system might work best?Your company wants to create asmartphone application that identifies plants using the phone’s camera. The companypurchases millions of digital images of plants labeled with the species names. You use thisinitial batch of images to train your artificial neural network, What type of machinelearning are you using for your network?

• No, these systems are incapable of breaking the law.
• No, these systems may be thought of as experiencing events that it hasn’t experienced.
• Yes, reciting what others have written about sunsets is plagiarism.
• It’s unclear, so there needs to be a new measure of authenticity.

牛津词典将抄袭定义为“采用他人的工作或想法并将其冒充为自己的行为”。如果你让 ChatGPT 描述一次日落，它会给你一个回应，但这些系统从未经历过日落，它唯一能回应的方式就是“把想法当作自己的想法”。这是否意味着这些生成式 Alsystems 是抄袭机器？您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周意见信，以表达您对行业头条新闻的看法。您创建了一些测试帖子，您注意到 ChatGpis 的日期错误，并且混淆了不同公司的首席执行官。你为什么要面对这个挑战？你的公司想要利用生成式人工智能来开发新的药物。该系统将分析所有现有的化合物，并尝试根据当前某些药物的成功开发新化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 不，这些系统不会违法
• 不，这些系统可能被认为正在经历它没有经历过的事件
• 是的，背诵别人写的关于日落的内容是抄袭。
• 目前尚不清楚，因此需要有新的真实性衡量标准

抄袭的概念植根于人类的意图和原创性。像 ChatGPT 这样的人工智能模型没有意图或意识。他们根据接受过训练的数据模式生成内容。虽然输出可能类似于现有内容，但尚不清楚这是否可以归类为传统意义上的抄袭。随着人工智能的不断发展，可能需要新的措施和定义来解决这些问题。

第 33 题

Typically, what are the three layers of an artificial neural network?You work in themarketing department for a large company, and you’d like to create a weekly opinion letterusing ChatGPT to give your take on the top news in your industry. You create a few testposts, and you notice that ChatGPT is getting the dates wrong and is mixing up the CEOs ofdifferent companies. Why are you running into this challenge?Your company wants to usegenerative Al to come up with new pharmaceuticals. This system will analyze all existingchemical compounds and try to develop new compounds based on the success of some ofyour current pharmaceuticals. This system will require a lot of custom programming andaccess to your proprietary data sets. What type of generative Al system might work best?Your company wants to create a smartphone application that identifies plants using thephone’s camera. The company purchases millions of digital images of plants labeled withthe species names. You use this initial batch of images to train your artificial neuralnetwork, What type of machine learning are you using for your network?

• the supervised layer, the unsupervised layers, and the reinforcement layer
• the artificial layer, the machine learning layer, and the data layer
• the input layer, many hidden layers, and the output layer
• the transaction layer, the generator layer, and the final layer

通常，人工神经网络分为哪三层？您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周意见信，以表达您对行业热门新闻的看法。您创建了一些测试帖子，然后您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要利用再生人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前的一些药物的成功开发新的化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 监督层、无监督层、加强层
• 人工层、机器学习层、数据层
• 输入层、许多隐藏层和输出层
• 交易层、生成器层和最终层

在典型的人工神经网络中，有一个输入层，将数据输入网络，一个或多个隐藏层，在其中进行计算并学习特征，以及一个输出层，在其中进行最终预测或分类

第 34 题

You’re trying to get better at prompt engineering, so you decided to try a new technique.You say, “Write a 500-word essay on large language models and hallucinations from theperspective of a computer science graduate student at a university.” What technique areyou using here?You work in the marketing department for a large company, and you’d liketo create a weekly opinion letter using ChatGPT to give your take on the top news in yourindustry. You create a few test posts, and you notice that ChatGPT is getting the dateswrong and is mixing up the CEOs of different companies. Why are you running into thischallenge?Your company wants to use generative Al to come up with new pharmaceuticals.This system will analyze all existing chemical compounds and try to develop newcompounds based on the success of some of your current pharmaceuticals. This system willrequire a lot of custom programming and access to your proprietary data sets. What typeof generative Al system might work best?Your company wants to create a smartphoneapplication that identifies plants using the phone’s camera. The company purchasesmilions of digital images of plants labeled with the species names. You use this initial batchof images to train your artificial neural network. What type of machine learning are youusing for your network?

• You are using role-playing to get more accurate responses.
• You are taking an adversarial approach to get both sides of the story.
• You are using a compression technique by limiting the results to 500 words.
• You’re brainstorming with the system about large language models and hallucinations.

你想在即时工程方面做得更好，所以你决定尝试一种新技术，你说，“从大学计算机科学研究生的角度写一篇关于大语言模型和幻觉的 500 字论文。”您在这里使用什么技术？您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周意见信，以表达您对行业热门新闻的看法。您创建了一些测试帖子，然后您注意到 ChatGPT 的日期有误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要使用生成式人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试在你当前的一些药物的成功基础上开发新的化合物。该系统需要大量自定义编程和访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，可以使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络，您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 您正在使用角色扮演来获得更准确的答复。
• 你正在采取对抗性的方法来了解故事的双方。
• 您正在使用压缩技术，将结果限制为 500 个字。
• 您正在与系统就大型语言模型和幻觉进行头脑风暴。

通过指定“大学计算机科学研究生”的观点，您可以引导模型生成与该特定观点一致的内容，这是角色扮演的一种形式。

第 35 题

You are going to use machine learning to try and do a better job predicting the weather. Tostart out, you just want to classify two weather events: “rain” or “not rain.” What stepswould you take to build this system?You work in the marketing department for a largecompany, and you’d like to create a weekly opinion letter using ChatGPT to give your takeon the top news in your industry. You create a few test posts, and you notice that ChatGPTis getting the dates wrong and is mixing up the CEOs of different companies. Why are yourunning into this challenge?Your company wants to use generative Al to come up with newpharmaceuticals. This system will analyze all existing chemical compounds and try todevelop new compounds based on the success of some of your current pharmaceuticals.This system will require a lot of custom programming and access to your proprietary datasets. What type of generative Al system might work best?Your company wants to create asmartphone application that identifies plants using the phone’s camera. The companypurchases millions of digital images of plants labeled with the species names. You use thisinitial batch of images to train your artificial neural network. What type of machinelearning are you using for your network?

• Use reinforcement learning to allow the machine to create rewards for itself based on howwell it predicted the weather.
• Use a linear regression to show the trend line from “not rain” to “rain.
• Find labeled weather data, create a small training set of that data, and that set aside moredata for the test set.
• Input all the labeled weather data and allow the system to create its own clusters based onwhat it sees in the data.

您将使用机器学习来尝试更好地预测天气。首先，您只想对两种天气事件进行分类：“下雨”或“不下雨”。您将采取什么步骤来构建这个系统？您在一家大公司的营销部门工作，您希望使用 ChatGpT 创建每周意见信，以表达您对行业头条新闻的看法。您创建了一些测试帖子，您注意到 ChatGpis 的日期错误，并且混淆了不同公司的首席执行官。你为什么要面对这个挑战？你的公司想要利用生成式人工智能来开发新的药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量自定义编程并访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 使用强化学习让机器根据天气预测的效果为自己创造奖励。
• 使用线性回归显示从“不下雨”到“下雨”的趋势线。
• 查找标记的天气数据，创建该数据的小型训练集，并为测试集留出更多数据。
• 输入所有标记的天气数据，并允许系统根据在数据中看到的内容创建自己的集群。

第 36 题

What’s one of the key dangers for organizations that over rely on generative Al systems?You work in the marketing department for a large company, and you’d like to create aweekly opinion letter using ChatGPT to give your take on the top news in your industry.You create a few test posts, and you notice that ChatGPT is getting the dates wrong and ismixing up the CEOs of different companies. Why are you running into this challenge?Yourcompany wants to use generative Al to come up with new pharmaceuticals. This system willanalyze all existing chemical compounds and try to develop new compounds based on thesuccess of some of your current pharmaceuticals. This system will require a lot of customprogramming and access to your proprietary data sets. What type of generative Al systemmight work best?Your company wants to create a smartphone application that identifiesplants using the phone’s camera. The company purchases millions of digital images ofplants labeled with the species names. You use this initial batch of images to train yourartificial neural network, What type of machine learning are you using for your network?

• Generative Al systems might make key decisions about who works for the company.
• They will regenerate the same material without any spark of creativity.
• Generative Al systems will start to run these organizations with little to no human oversight.
• Your employees might resign if they feel that your system is in danger of replacing theirlivelihood.

对于过度依赖生成式 AI 系统的组织来说，主要危险之一是什么？您在一家大公司的营销部门工作，您希望使用 ChatGpT 创建每周意见书，以表达您对行业热门新闻的看法您创建了一些测试帖子，然后您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。你为什么会遇到这个挑战？你的公司想要使用生成式人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前的一些药物的成功开发新的化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 生成式人工智能系统可能会就谁为公司工作做出关键决策。
• 他们将重新生成相同的材料，但没有任何创造力的火花。
• 生成式人工智能系统将开始在几乎没有人为监督的情况下运行这些组织
• 如果您的员工认为您的系统有取代他们生计的危险，他们可能会辞职。

过度依赖人工智能，尤其是在决策过程中，可能会导致在没有人类判断的情况下做出决策，从而可能导致不可预见的后果或道德困境。

第 37 题

How does a reasoning engine work?You work in the marketing department for a largecompany, and you’d like to create a weekly opinion letter using ChatGPT to give your takeon the top news in your industry. You create a few test posts, and you notice that ChatGPTis getting the dates wrong and is mixing up the CEOs of different companies. Why are yourunning into this challenge?Your company wants to use generative Al to come up with newpharmaceuticals. This system will analyze all existing chemical compounds and try todevelop new compounds based on the success of some of your current pharmaceuticals.This system will require a lot of custom programming and access to your proprietary datasets. What type of generative Al system might work best?Your company wants to create asmartphone application that identifies plants using the phone’s camera. The companypurchases millions of digital images of plants labeled with the species names. You use thisinitial batch of images to train your artificial neural network. What type of machinelearning are you using for your network?

• It draws conclusions, makes decisions, summarizes information, and solves problems basedon available data.
• Ilt’s a way for search engines to crawl, index, and rank new content so that it’s always freshdata and able to solve real problems.
• It networks together several search engines so that users always have access to good content.
• It’s a way for computer scientists to optimize server code in a hosted reason repository.

推理引擎如何工作？您在一家大公司的营销部门工作，您希望使用 ChatGPT 创建每周意见信，以表达您对行业头条新闻的看法。您创建了一些测试帖子，您注意到 ChatGpTi 的日期错误，并且混淆了不同公司的首席执行官。你为什么要面对这个挑战？你的公司想要利用生成式人工智能来开发新的药物。该系统将分析所有现有的化合物，并尝试根据您当前某些药物的成功开发新化合物。该系统将需要大量自定义编程并访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器收益？

下面哪个选项是正确的

• 它根据可用数据得出结论、做出决策、总结信息并解决问题。
• 这是搜索引擎抓取、索引和排名新内容的一种方式，以便它始终是新鲜数据并能够解决实际问题。
• 它将多个搜索引擎联网，以便用户始终能够访问优质内容
• 这是计算机科学家在托管原因存储库中优化服务器代码的一种方法

推理引擎旨在逻辑地处理信息并从中得出结论，类似于人类推理的工作方式，但以结构化和基于规则的方式。

第 38 题

What is a good description of how a machine learning system operates?You work in themarketing department for a large company, and you’d like to create a weekly opinion letterusing ChatGPT to give your take on the top news in your industry. You create a few testposts, and you notice that ChatGPT is getting the dates wrong and is mixing up the CEOs ofdifferent companies. Why are you running into this challenge?Your company wants to usegenerative Al to come up with new pharmaceuticals. This system will analyze all existingchemical compounds and try to develop new compounds based on the success of some otyour current pharmaceuticals. This system will require a lot of custom programming andaccess to your proprietary data sets. What type of generative Al system might work best?Your company wants to create a smartphone application that identifies plants using thephone’s camera. The company purchases millions of digital images of plants labeled withthe species names. You use this initial batch of images to train your artificial neuralnetwork. What type of machine learning are you using for your network?

• An Al system learns in a way that is consistent with its preprogrammed responses.
• A system “learns” by observing patterns in massive datasets.
• An Al system generates content as opposed to just classifying existing data.
• A system achieves artificial general intelligence by collating responses from experts in every
• field.

关于机器学习系统如何运作的最佳描述是什么？您在一家大公司的营销部门工作，您希望使用 ChatGpT 创建每周意见信，以表达您对行业热门新闻的看法。您创建了一些测试帖子，然后您注意到 ChatGPT 的日期错误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要利用再生人工智能来研制新的药物。该系统将分析所有现有的化合物，并尝试根据您当前的一些药物的成功开发新的化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？/我们公司想要创建一款智能手机应用程序，可以使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 人工智能系统以与其预编程响应一致的方式进行学习。
• 系统通过观察海量数据集中的模式来“学习”
• 人工智能系统生成内容，而不是仅仅对现有数据进行分类。
• 系统通过整理各个领域专家的反应来实现通用人工智能。

机器学习模型在大型数据集上进行训练以识别模式，并使用这些模式对新的、未见过的数据进行预测或决策。

第 39 题

You manage a radiology department in a large hospital. Your hospital has millions ofcomputed tomography (CT) images. You want to create a system where once someone getsa CT scan, the system will immediately check for anomalies. That way it can be sent forreview by a senior radiologist. Which generative Al system might work best for thisapproach?You work in the marketing department for a large company, and you’d like tocreate a weekly opinion letter using ChatGPT to give your take on the top news in yourindustry. You create a few test posts, and you notice that ChatGPT is getting the dateswrong and is mixing up the CEOs of different companies. Why are you running into thischallenge?Your company wants to use generative Al to come up with new pharmaceuticals.This system will analyze all existing chemical compounds and try to develop newcompounds based on the success of some of your current pharmaceuticals. This system willrequire a lot of custom programming and access to your proprietary data sets. What typeof generative Al system might work best?Your company wants to create a smartphoneapplication that identifies plants using the phone’s camera. The company purchasesmillions of digital images of plants labeled with the species names. You use this initial batchof images to train your artificial neural network, What type of machine learning are youusing for your network?

• a generative autoencoding network (GAN)
• a variational autoencoder (VAEs)
• an adversarial autoencoder (AAA)
• a flexible learning encoding X-ray (FLEX)

您在一家大医院管理放射科。您的医院拥有数百万张计算机断层扫描 (CT) 图像。您想要创建一个系统，一旦有人接受 CT 扫描，系统就会立即检查是否有异常情况。这样就可以将其发送给高级放射科医生进行审查。哪种生成式人工智能系统最适合这种方法？您在一家大公司的营销部门工作，您希望使用 ChatGpT 创建每周意见信，以表达您对行业热门新闻的看法。您创建了一些测试帖子，然后您注意到 ChatGpT 的日期有误，并且混淆了不同公司的首席执行官。为什么你会遇到这个挑战？你的公司想要使用生成式人工智能来开发新的药物。该系统将分析所有现有的化合物，并尝试根据你当前的一些药物的成功开发新的化合物。该系统将需要大量的自定义编程和访问您的专有数据集。哪种类型的生成式人工智能系统效果最好？您的公司想要创建一款智能手机应用程序，可以使用手机的摄像头识别植物。该公司购买了数百万张标有物种名称的植物数字图像。您使用这批初始图像来训练您的人工神经网络。您的网络使用什么类型的机器学习？

下面哪个选项是正确的

• 生成自动编码网络（GAN）
• 变分自动编码器 (VAE)
• 对抗性自动编码器（AAA）
• 灵活的学习编码X射线（FLEX）

VAE 是一种生成模型，对于异常检测特别有用。它们可以被训练来重建输入数据，并且可以根据重建误差来检测异常。如果 CT 扫描与 VAE 预期的结果（基于其训练数据）显着不同，则可以将其标记为异常。

第 40 题

What is an Al model?

• a model is an example of how a generative Al application can be used
• a model is a set of algorithms that have been trained on a specific dataset
• a model is a tool for writing and running code

什么是AI模型？

• 模型是一个展示生成式AI应用如何被使用的例子。
• 模型是一套已经在特定数据集上训练过的算法。
• 模型是一个用于编写和运行代码的工具。

第 41 题

What is a key benefit of using generative Al for repetitive or computational tasks?

• It increases the flexibility and adaptability of the inputs.
• It allows humans to focus on more creative and strategic activities.
• It reduces the risk of errors and biases.
• It improves the quality and efficiency of the outputs.

使用生成式AI处理重复性或计算任务的一个关键好处是什么？

• 它增加了输入的灵活性和适应性。
• 它允许人类专注于更多创造性和战略性的活动。
• 它减少了错误和偏见的风险。
• 它提高了输出的质量和效率。

第 42 题

What is the difference between generative Al and other types of Al that generate content?

• Generative Al is a subset of reactive machines that generate content in response to stimuli.
• Generative Al’s primary function is to create content.
• Generative Al uses unsupervised learning to generate content without preexisting data.
• Generative Al can generate content in any domain or format

生成式AI与其他生成内容的AI类型之间的区别是什么？

• 生成式AI是反应机器的一个子集，它根据刺激生成内容。
• 生成式AI的主要功能是创建内容。
• 生成式AI使用无监督学习，在没有预先存在的数据的情况下生成内容。
• 生成式AI可以在任何领域或格式中生成内容。

第 43 题

Do you need to have a technical Al background in order to start a generative Al venture?

• Yes, because it is hyper complex, Several generative Al models and papers comes out every
  day and in order to work in generative Al, you need to be able to write generative Al
  algorithms and you need to be able to train your own datasets.
• No. Generative Al is so advanced that the code is writing itself, You almost do not need to
  hire anyone, you can do everything, including programming to product development on you
  own.
• No. As a business leader with no technical background, you can either make partnerships with
  generative Al research institutions, or you can use open-source models in your next business
  endeavor. And as a hobbyist, a maker, or a creative, you can make use of a generative Al services
  that are free or are accessible with a modest fee.

为了开始一个生成式AI企业，你需要拥有技术背景吗？

• 是的，因为它极其复杂。每天都会出现几个生成式AI模型和论文，为了在生成式AI领域工作，你需要能够编写生成式AI算法，并且需要能够训练你自己的数据集。
• 不。生成式AI如此先进，以至于代码正在自我编写。你几乎不需要雇佣任何人，你可以自己做所有事情，包括编程到产品开发。
• 不。作为一个没有技术背景的商业领袖，你可以与生成式AI研究机构建立合作伙伴关系，或者你可以在你下一个商业努力中使用开源模型。作为一个业余爱好者、制造者或创意人士，你可以使用免费的或以适度费用可获得的生成式AI服务。

第 44 题

What does the term open source mean in the context of generative Al models?

• It means that the models are transparent and explainable to human users.
• It means that the models are generated by crowdsourcing data from online users.
• It means that the models are funded by the public sector and non-profit organizations.
• lt means that the models are publicly available for anyone to use and modify.

在生成式AI模型的背景下，“开源”这个术语意味着什么？

• 它意味着模型对人类用户是透明的并且可以解释的。
• 它意味着模型是通过从在线用户那里众包数据生成的。
• 它意味着模型是由公共部门和非盈利组织资助的。
• 它意味着模型可公开供任何人使用和修改。

第 45 题

How can Variational Autoencoders (VAEs) be used in anomaly detection?

• The natural language processing of VAEs help them compute complex information. With their large scale transformer architecture, they can quickly process language-based information.
• VAEs are trained with large datasets and they have the capability to future predict anomalies by analyzing the behaviors of a production systems.
• VAEs can be trained on a dataset of normal data, and later on be used to identify instances that deviate from the normal data.

变分自编码器（VAEs）如何用于异常检测？

• VAEs的自然语言处理帮助它们计算复杂信息。通过它们的大规模变换器架构，它们可以快速处理基于语言的信息。
• VAEs通过大型数据集进行训练，它们具有通过分析生产系统的行为来预测未来异常的能力。
• VAEs可以在正常数据的数据集上进行训练，随后用来识别偏离正常数据的实例。

第 46 题

What is one real-world application of Variational Autoencoders (VAE) in anomaly detection?

• Creating realistic audio samples for use in music production.
• Identifying patterns in social media activity to predict consumer behavior.
• Detecting defects in industrial quality control by identifying images of products that deviate from a dataset of normal products.
• Generating synthetic fraudulent transactions to train financial fraud detection models.

变分自编码器（VAE）在异常检测中的一个现实世界应用是什么？

• 创建逼真的音频样本，用于音乐制作。
• 识别社交媒体活动中的模式，以预测消费者行为。
• 在工业质量控制中检测缺陷，通过识别与正常产品数据集偏离的产品图像。
• 生成合成的欺诈交易，以训练金融欺诈检测模型。

第 47 题

What is the most notable functionality of Natural Language models like Chat GPT?

• its capability to write hyper creative text for a variety of contents such as books, slogans, and scripts
• its large scale capability to generate human-like text
• its ability to understand the meaning of the text it generates
• its capability to influence people to make decisions for them

像Chat GPT这样的自然语言模型最显著的功能是什么？

• 它能够为书籍、口号和剧本等多种内容创作极具创意的文本。
• 它具有大规模生成类似人类文本的能力。
• 它能够理解自己生成的文本的含义。
• 它能够影响人们为他们做决定。

第 48 题

What is GPT and why has it become notable in the field of natural language processing?

• GPT is a language model developed by OpenAl that can take in a prompt and generate text based on it, making it useful for a multitude of tasks.
• GPT is a language model developed by OpenAl that uses a transformer architecture to generate human-like texts.
• GPT is a language model developed by Google that uses a pre-training technique to improve its performance on task-specific datasets.
• GPT is a language model developed by Microsoft that suggests code and entire functions inreal-time to its users.

GPT是什么，以及为什么它在自然语言处理领域变得突出？

• GPT是由OpenAI开发的一种语言模型，能够接受一个提示并基于它生成文本，使其对多种任务都很有用。
• GPT是由OpenAI开发的一种语言模型，使用变换器架构生成类似人类的文本。
• GPT是由Google开发的一种语言模型，使用预训练技术来提高其在特定任务数据集上的表现。
• GPT是由Microsoft开发的一种语言模型，能够实时为用户推荐代码和完整函数。

第 49 题

What are Midjourney, DALL-E, and Stable Diffusion, and which industries are their early adopters?

• They are large language models models. Chatbots, search engines, and customer service are the primary industry adopters.
• They are 3D asset generation companies, Their generated outcomes are used to design clothes objects, CGl VFX, and are helping filmmakers quickly generate 3D environments.
• They are primary text-to-image generation services and models. Art, filmmaking, fashion, and marketing are the first industries to widely adopt their use.

Midjourney、DALL-E和Stable Diffusion是什么，以及它们的早期采用者是哪些行业？

• 它们是大型语言模型。聊天机器人、搜索引擎和客户服务是主要的行业采用者。
• 它们是3D资产生成公司，它们生成的结果被用来设计服装物品、CGI特效，并且帮助电影制作人快速生成3D环境。
• 它们主要是文本到图像生成服务和模型。艺术、电影制作、时尚和营销是最先广泛采用它们的行业。

第 50 题

Why might marketers use text-to-image models in the creative process?

• They can efficiently suggest garments for customers based on their fashion style.
• They can be more efficient to use and provide a unique look and feel.
• They have the ability to generate 3D worlds for films more efficiently.

为什么营销人员在创意过程中可能会使用文本到图像模型？

• 它们能够根据顾客的时尚风格高效地推荐服装。
• 它们的使用更加高效，并能提供独特的外观和感觉。
• 它们具有更高效地为电影生成3D世界的能力。

第 51 题

How does a GAN network improve its ability to generate better content?

• The generator and discriminator parts of the network work together in harmony to challenge and trick the user in identifying which outcomes are “real” and which are “synthetic”.
• The generator and discriminator parts of the network work together in a competition to improve the generator’s ability to create realistic data.
• The user writes text to generate content and the networks learns to improve itself each time it is being used.

GAN网络如何提高其生成更好内容的能力？

• 网络的生成器和鉴别器部分和谐地一起工作，挑战和欺骗用户识别哪些结果是“真实的”和哪些是“合成的”。
• 网络的生成器和鉴别器部分在一场竞争中共同工作，以提高生成器创建逼真数据的能力。
• 用户编写文本来生成内容，网络每被使用一次就学习提升自己。

第 52 题

What is the purpose of the discriminator in a GAN model?

• to input one type of data and output the same type of data
• to create realistic data for training models
• to generate synthetic versions of fraudulent transactions for financial fraud detection
• to evaluate the data created by the generator and give feedback on how to improve the next iteration

在GAN模型中，鉴别器的目的是什么？

• 输入一种类型的数据并输出相同类型的数据
• 为训练模型创建逼真的数据
• 生成合成版本的欺诈交易，用于金融欺诈检测
• 评估生成器创建的数据，并就如何改进下一次迭代提供反馈

第 53 题

What will be the main benefit of generative Al in the next years?

• automate repetitive tasks and liberate humanity from dull, dirty, difficult, or dangerous jobs
• optimize supply chains for corporations to save excess spending
• increase security through Blockchain and distributed ledger technology

在接下来的几年里，生成式AI的主要好处将是什么？

• 自动化重复任务，解放人类免受枯燥、肮脏、艰难或危险的工作
• 为企业优化供应链，节省过度支出
• 通过区块链和分布式账本技术增加安全性

第 54 题

In considering the future implications of generative Al on the job market, what skill set is essential for individuals seeking to thrive in this evolving landscape?

• mastery of machine learning algorithms and programming languages used in generative Al
• specialized technical skills in operating and managing automated systems
• development of unique personal emotional, interpersonal, and creative abilities
• proficiency in utilizing existing creative tools and software for content generation

考虑到生成式AI对就业市场未来的影响，寻求在这个不断发展的领域中蓬勃发展的个人需要哪些技能？

• 精通生成式AI中使用的机器学习算法和编程语言
• 在操作和管理自动化系统方面的专业技能
• 发展独特的个人情感、人际交往和创造能力
• 熟练利用现有的创意工具和软件进行内容生成

第 55 题

What are the top moral and executive skill sets required when working with generative Al?

• Transparency, fairness, empathy and responsibility. Approach production and operations with caution, always asking, “Who is benefiting?” from our generative Al solution.
• Lobbying and diplomacy.
• A rich technical background, especially in machine learning and natural language processing coupled with excellent leadership skills.

在使用生成式AI工作时，需要哪些最重要的道德和执行技能？

• 透明度、公平性、同理心和责任感。小心谨慎地处理生产和操作过程，始终询问：“谁从我们的生成式AI解决方案中受益？”
• 游说和外交技巧。
• 丰富的技术背景，特别是在机器学习和自然语言处理方面，加上出色的领导能力。

第 56 题

When considering the integration of generative Al tools in business operations, what is the primary emphasis regarding the role of executive leadership and organizational strategy?

• encouraging blind reliance on Al-generated content without considering ethical implications or human oversight
• prioritizing human-centered approaches, ethical considerations, and maintaining human control over Al-generated content
• advocating for unrestricted reliance on generative Al outcomes to streamline decision-making processes
• establishing rigid guidelines that prioritize algorithm-generated content over human decision-making

在考虑将生成式AI工具整合到商业运营中时，对于高层领导和组织战略的角色，主要强调什么？

• 鼓励盲目依赖AI生成的内容，而不考虑伦理影响或人类监督
• 优先考虑以人为本的方法、伦理考量，并保持对AI生成内容的人类控制
• 提倡对生成式AI成果的无限制依赖，以简化决策过程
• 建立优先考虑算法生成内容而非人类决策的严格指导方针

第 57 题

How should we think of the relationship between humans and generative Al?

• Human input and creativity will work in conjunction with Al to produce meaningful progress
• Al is necessary to maintain objectivity and eliminate human bias.
• Al will become singularly responsible for decision-making and creativity.
• Al will likely devalue human contributions and creativity.

我们应该如何看待人类与生成式AI之间的关系？

• 人类的输入和创造力将与AI一起工作，以产生有意义的进步。
• AI是保持客观性和消除人类偏见所必需的。
• AI将独立负责决策和创造性工作。
• AI可能会贬低人类的贡献和创造力。

第 58 题

What is the goal of the continuous crawling process of a search engine?

• to discover new or updated webpages
• to keep the search engine’s index up-to-date
• to store and organize webpage information

搜索引擎持续爬取过程的目标是什么？

• 发现新的或已更新的网页
• 保持搜索引擎索引的最新状态
• 存储和组织网页信息

第 59 题

When a user enters a query, what does the reasoning engine strive to provide?

• a relevant, informative text response using human-like speech
• a brief bulleted list summarizing the main points of the query
• a ranked list of webpage results

当用户输入一个查询时，推理引擎努力提供什么？

• 使用类似人类的语言提供相关、信息丰富的文本回应
• 一个简短的项目符号列表，总结查询的主要点
• 一个排名的网页结果列表

因为推理引擎处理并理解人类语言，它们能够使用类似人类的语言提供相关、经过良好推理的、信息丰富的回应。

第 60 题

When might a search engine be a superior tool to a reasoning engine?

• when you’ d like to read further about a subject across a collection of different sources-but not necessarily when you want to ask deeper questions
• when you want to have an intelligent conversation with an Al chatbot
• when you want a customized, well-reasoned answer generated with human-like speech

搜索引擎何时可能是比推理引擎更优越的工具？

• 当你想要通过一系列不同的来源更深入地了解一个主题时——但不一定是当你想要提出更深入的问题时
• 当你想要与一个AI聊天机器人进行智能对话时
• 当你想要一个定制化的、经过良好推理的、用类似人类的语言生成的答案时

第 61 题

Which is not a main function of a search engine?

• transforming
• crawling
• indexing
• ranking

哪项不是搜索引擎的主要功能？

• 转换
• 爬行
• 索引
• 排名

搜索引擎的主要功能是爬行、索引和排名。

第 62 题

What is the most important benefit that the synergy between modern search engines and reasoning engines provides, as far as confidence in the results?

• avoiding time wasted on unrelated results
• being able to use both engines simultaneously on a single platform
• verifying and validating search results

就对结果的信心而言，现代搜索引擎和推理引擎之间的协同作用提供的最重要好处是什么？

• 避免浪费时间在无关的结果上
• 能够在单一平台上同时使用这两种引擎
• 验证和确认搜索结果

第 63 题

How can a user best combine a search engine and a reasoning engine to find information about an unknown topic?

• Use the reasoning engine to get a summary of the topic, and then use the search engine for more detailed information.
• Use one engine to find general information, and then use the other engine to find disputing information on that topic.
• Use the search engine to find basic information, and then use the reasoning engine for a deeper dive.

用户如何最好地结合搜索引擎和推理引擎来查找关于一个未知主题的信息？

• 使用推理引擎获得该主题的概述，然后使用搜索引擎寻找更详细的信息。
• 使用一个引擎找到一般信息，然后使用另一个引擎找到关于那个主题的争议信息。
• 使用搜索引擎找到基本信息，然后使用推理引擎进行更深入的探索。

第 64 题

How does a reasoning engine’s ability to understand and interpret language provide the greatest advantage over a search engine?

• lt can provide a collection of additional ranked results on the topic requested.
• It can have an actual conversation with the user
• It can mediate conflicting information between different machine learning models.

推理引擎理解和解释语言的能力与搜索引擎相比，提供了最大的优势是什么？

• 它可以提供所请求主题的额外排名结果集。
• 它可以与用户进行实际对话。
• 它可以在不同机器学习模型之间调解冲突信息。

第 65 题

How are reasoning engines an improvement over search engines when it comes to entering what you are looking for?

• They can understand your intent and not just the words you used.
• They can provide a direct, ranked answer to your request.
• They can suggest alternative search queries to use if they sense yours needs improvement

当输入你正在寻找的内容时，推理引擎如何比搜索引擎有所改进？

• 它们可以理解你的意图，而不仅仅是你使用的词汇。
• 它们可以直接提供一个排名的答案来回应你的请求。
• 如果它们感觉到你的查询需要改进，它们可以建议使用其他的搜索查询。

第 66 题

How do human supervisors assist in training a reasoning engine?

• Human supervisors perform monthly audits on reasoning engine responses to ensure the Al is doing a suitable job.
• In early training phases, human supervisors oversee the process, guiding the model towards accurate responses and contributing to its knowledge development.
• Human supervisors monitor every user query to make sure the reasoning engine creates aCCl responses

人类监督者如何协助训练一个推理引擎？

• 人类监督者对推理引擎的响应进行月度审计，以确保AI工作适当。
• 在早期训练阶段，人类监督者监督这一过程，引导模型向准确的响应发展，并为其知识发展做出贡献。
• 人类监督者监控每一个用户查询，以确保推理引擎创建准确的回应。

第 67 题

True or False: Reasoning engines are an all-knowing source of truth and should be trusted implicitly.

• FALSE
• TRUE

真或假：推理引擎是一个全知的真理来源，应该被隐含地信任。

• 假
• 真

第 68 题

When a user enters a query, what does the reasoning engine strive to provide?

• a relevant, informative text response using human-like speech
• a brief bulleted list summarizing the main points of the query
• a ranked list of webpage results

当用户输入一个查询时，推理引擎努力提供什么？

• 使用类似人类的语言的相关、信息丰富的文本回应
• 一个简短的项目符号列表，总结查询的主要点
• 一个排名的网页结果列表

因为推理引擎处理并理解人类语言，它们能够使用类似人类的语言提供相关、经过良好推理的、信息丰富的回应。

第 69 题

When might a search engine be a superior tool to a reasoning engine?

• when you’ d like to read further about a subject across a collection of different sources-but not necessarily when you want to ask deeper questions
• when you want to have an intelligent conversation with an Al chatbot
• when you want a customized, well-reasoned answer generated with human-like speech

搜索引擎何时可能是比推理引擎更优越的工具？

• 当你想要通过一系列不同的来源更深入地了解一个主题时——但不一定是当你想要提出更深入的问题时
• 当你想要与一个AI聊天机器人进行智能对话时
• 当你想要一个定制化的、经过良好推理的、用类似人类的语言生成的答案时

第 70 题

If your reasoning engine response is problematic (i.e., inaccurate, discriminatory
limited in view, etc.) what should you do?

• Report the problematic response to the FCc
• Give up and seek other research methods; the reasoning engine is unlikely to produce a valid result.
• Continue iterating. Keep regenerating and refining the prompt to get a more accurate, better result.

如果你的推理引擎响应存在问题（即不准确、有歧视性、观点有限等），你应该怎么做？

• 将问题响应报告给FCC
• 放弃并寻求其他研究方法；推理引擎不太可能产生有效的结果。
• 继续迭代。不断重新生成和精炼提示，以获得更准确、更好的结果。

第 71 题

In prompt engineering, what is one-shot or few-shot learning?

• lt refers to the style of answer you want to target in your response.
• It refers to how many chances you give the reasoning engine to get the right answer.
• It refers to how much instruction you provide in order to quide the answer. This may involve including examples of what a “correct” answer may look like.

在提示工程中，什么是单次学习或少次学习？

• 它指的是你想要在回应中定位的答案风格。
• 它指的是你给推理引擎多少次机会来得到正确答案。
• 它指的是你提供多少指令以引导答案。这可能包括包含一个“正确”答案可能看起来像的例子。

第 72 题

In most instances, how should you craft your prompts?

• Use clear language with proper grammar.
• Use conversational and informal language.
• Use jargon common to the industry.

在大多数情况下，你应该如何构建你的提示？

• 使用清晰的语言和正确的语法。
• 使用会话式和非正式语言。
• 使用行业常见的术语。

第 73 题

Why is the iteration process necessary when you use a reasoning engine?

• You want to use a combination of both keywords and regular language prompts.
• You want to keep honing your prompt to get more and better results.
• You want to avoid creating multiple prompts when one qood prompt will suffice

为什么在使用推理引擎时，迭代过程是必要的？

• 你想要同时使用关键词和常规语言提示的组合。
• 你想要不断完善你的提示，以获得更多更好的结果。
• 你想要避免在一个好的提示就足够的情况下创建多个提示。

第 74 题

Riva considers herself a prompt engineer, What does this mean?

• She can use coding to create better prompts than the prompts she creates with regular anguage.
• She can create a prompt with samples of her question and the answers that she would like to see.
• She has created a number of prompts to the point where she is an expert on crafting prompts.

Riva认为自己是一名提示工程师，这意味着什么？

• 她可以使用编程来创建比她用常规语言创建的提示更好的提示。
• 她可以创建一个带有她问题及其希望看到的答案样本的提示。
• 她已经创建了许多提示，以至于她成为了制作提示的专家。

第 75 题

What is the following creative type of prompt known as: “lmagine you’re the manager of a small botique video editing company. What are 10 innovative marketing ideas that could attract new business?

• Debate-style
• Role play
• Analogy

以下这种创意类型的提示被称为什么：“想象你是一家小型精品视频编辑公司的经理。你能想出10个创新的营销点子来吸引新生意吗？

• 辩论式
• 角色扮演
• 类比

第 76 题

True or False: Reasoning engines are an all-knowing source of truth and should be trusted implicitly.

• FALSE
• TRUE

真或假：推理引擎是一个全知的真理来源，应该被绝对信任。

• 假
• 真

第 77 题

Excel Copilot can only be used if ____.

• you are working with Excel on Window
• you are working with Excel on the web
• the file you are working on is stored on OneDrive, SharePoint, or Teams

Excel Copilot只能在____的情况下使用。

• 你在Windows上使用Excel
• 你在网页上使用Excel
• 你正在工作的文件存储在OneDrive、SharePoint或Teams上

当使用Word或PowerPoint工作时，Copilot可以用于存储在你电脑上的文件。但是，在使用Excel时，Copilot只适用于存储在微软的云存储工具之一上的文件。

第 78 题

Excel Copilot can only analyze data in a spreadsheet if ____.

• the sheet contains 5 or more columns of data.
• any cells containing numbers are formatted as currency.
• You select the cells before opening the Copilot panel.
• the cells you want to analyze are formatted as a table.

Excel Copilot只能在电子表格中分析数据，如果____。

• 表格包含5列或更多列数据。
• 任何包含数字的单元格都格式化为货币。
• 在打开Copilot面板之前，你已经选择了单元格。
• 你想分析的单元格被格式化为表格。

第 79 题

If you join a Teams meeting that’s already in progress, you can ask Copilot to summarize the topics that were discussed that you may have missed. This will only work if ____.

• somebody in the meeting clicks the Copilot option and turns on the transcript feature
• each person in the meeting gives their permission
• there are 5 or more people in the meeting

如果你加入了一个已经开始的Teams会议，你可以要求Copilot总结你可能错过的讨论话题。这只有在____的情况下才会起作用。

• 会议中的某人点击Copilot选项并开启了转录功能
• 会议中的每个人给予了他们的许可
• 会议中有5人或更多人

第 80 题

PowerPoint Copilot can create a new presentation based on your written prompts or ____.

• information in an email message
• information in an existing document.
• information from a Teams meeting.

PowerPoint Copilot可以根据你的书面提示或____创建一个新的演示文稿。

• 邮件中的信息
• 现有文档中的信息。
• Teams会议中的信息。

第 81 题

What is a common reason the summary option may not be available in Outlook?

• You are using Outlook on the web.
• The message you have selected is very short and does not have enough information to summarize.
• You are using Outlook on the mac.

在Outlook中，摘要选项可能不可用的一个常见原因是什么？

• 你在网页上使用Outlook。
• 你选择的消息非常短，没有足够的信息进行摘要。
• 你在Mac上使用Outlook。

第 82 题

If you want to ask Word Copilot to use information from a separate document, that
referenced document must be ____.

• less than 90 days old
• shared in Teams or stored on OneDrive, SharePoint
• stored in the same folder as the currently open document
• formatted according to Copilot standards

如果你想要请求Word Copilot使用来自另一个文档的信息，那个参考文档必须是____。

• 不超过90天老
• 在Teams共享或存储在OneDrive、SharePoint上
• 存储在与当前打开文档相同的文件夹中
• 按照Copilot标准格式化

你可以在计算机的存储驱动器上存储的文档中使用Copilot。但是，如果你想在发出Copilot请求时引用另一个文件，那个引用的文件必须存储在微软的一个云存储工具上。

第 83 题

What type of misused Al can give false advice, which is extremely dangerous in a situation such as providing medical advice?

• legal confusion
• deep fakes
• inaccurate chatbots

哪种类型的误用人工智能会给出错误的建议，在提供医疗建议等情况下极其危险？

• 法律混乱
• 深度伪造
• 不准确的聊天机器人

第 84 题

ABC Corp’s management uses Al to provide advice on selecting options for its marketing strategy, How can robust transparency help the company?

• It can assist in understanding the input-analysis-output process.
• It can identify the most responsible way to achieve goals.
• It can uncover bias in the Al input training data.

ABC公司的管理层使用AI来提供有关选择其营销策略选项的建议，健全的透明度如何帮助该公司？

• 它可以帮助理解输入-分析-输出过程。
• 它可以识别实现目标的最负责任的方式。
• 它可以揭示AI输入训练数据中的偏见。

第 85 题

Layla learns that her company’s Al-driven chatbot has been giving inaccurate and sometimes rude responses to customers” questions. How should Layla deal with this issue?

• Limit the number of employees who have access to the chatbot data set.
• Limit the chatbot to customer service questions and no other purpose.
• Limit the number of customers who can use the chatbot at any one time.

Layla得知她公司的AI驱动的聊天机器人对客户的问题给出了不准确甚至有时候是粗鲁的回应。Layla应该如何处理这个问题？

• 限制能够访问聊天机器人数据集的员工数量。
• 将聊天机器人限制在客户服务问题上，不用于其他目的。
• 限制任何时候可以使用聊天机器人的客户数量。

第 86 题

Ahmad is creating a technology team for a new project. When should the team get together to discuss ethical considerations?

• as ethical issues arise during the project
• before starting the project
• during the team member interview process

Ahmad 正在为一个新项目组建一个技术团队。团队应该什么时候聚在一起讨论伦理考量？

• 在项目进行中出现伦理问题时
• 在开始项目之前
• 在团队成员面试过程中

第 87 题

Why is it a good practice for the C-suite to develop, and perhaps even mandate, an AI training and education program for all employees?

• It democratizes decision-making around AI tools.
• It increases employee engagement and retention.
• It gives employees the responsibility for training on AI tools.

为什么对于 C 级高管来说，开发乃至可能强制要求所有员工参与人工智能培训和教育计划是一个好做法？

• 它使围绕 AI 工具的决策民主化。
• 它增加了员工的参与度和留存率。
• 它给予员工负责 AI 工具培训的责任。

第 88 题

In which situation are ethical considerations the responsibility of a board of directors?

• when it involves compliance with a regulatory agency
• when it involves day-to-day organizational activities
• when it involves organizational hiring and training

在哪种情况下，伦理考量是董事会的责任？

• 当涉及到遵守监管机构时
• 当涉及到日常组织活动时
• 当涉及到组织的招聘和培训时

第 89 题

How can a company best build customer trust, which can translate to a loyal customer base?

• Listen to customers before developing a new product.
• Involve customers in decisions.
• Share a transparent privacy policy.

公司如何最有效地建立顾客信任，进而转化为忠实的顾客基础？

• 在开发新产品之前听取顾客的意见。
• 让顾客参与决策。
• 分享一个透明的隐私政策。

第 90 题

Why, more than any other reason, should a company perform a privacy audit?

• so it can develop a company-wide privacy policy
• so it can identify private customer information in its possession
• so it can establish guidelines for private information access

为什么，比其他任何原因都更重要，公司应该进行隐私审计？

• 以便它能够制定一个全公司范围的隐私政策
• 以便它能够识别其拥有的私人顾客信息
• 以便它能够建立私人信息访问指南

第 91 题

Why did some of the earliest artificial intelligence systems focus on board games such as checkers and chess?

• It’s easiest to make a computer system seem intelligent when it’s working with set rules and patterns.
• Because early computer scientists didn’t want the system to seem too intelligent.
• Board games gave computer systems access to huge amounts of data which allowed the machine to learn new things.
• Board games were an easy way to have computers create neural pathways.

为什么一些最早的人工智能系统专注于棋盘游戏，如跳棋和国际象棋？

• 当它按照固定的规则和模式工作时，让计算机系统看起来智能是最简单的。
• 因为早期的计算机科学家不想让系统看起来过于智能。
• 棋盘游戏让计算机系统能够接触到大量的数据，这使得机器能学到新东西。
• 棋盘游戏是让计算机创建神经路径的一种简单方式。

第 92 题

You’re a product manager who’s in charge of building a weak AI expert system that will give tax advice. You’re working with dozens of accountants who go through thousands of different taxpayer scenarios. When a customer asks a question, then the expert system will ask a follow-up question. It will do this until it makes a recommendation. What’s one of the biggest challenges with this system?

• The system could evolve into strong AI and develop a personality.
• Many people will be uncomfortable trusting a computer system with their taxes.
• There will be too many tax combinations for the experts to cover with one system.
• There aren’t enough tax experts to help develop scenarios for the system.

你是一个负责构建一个弱 AI 专家系统来提供税务建议的产品经理。你与数十位会计师合作，他们处理成千上万个不同的纳税人场景。当顾客提问时，专家系统会提出后续问题。它会这样做直到做出推荐。这个系统最大的挑战之一是什么？

• 系统可能会进化成强 AI 并发展出一种个性。
• 许多人会不习惯信任计算机系统处理他们的税务。
• 对于专家来说，一个系统要覆盖的税务组合太多了。
• 没有足够的税务专家帮助为系统开发场景。

第 93 题

Luella seeks medical attention for chest pains. A nurse uses an artificial intelligence program to diagnose the cause. Why is this system likely not really intelligent?

• The program can only be intelligent if the patient provides a complete medical history.
• The program only matches her symptoms to steps in a system an expert created.
• The program is only intelligent if a patient has been there before.

Luella 因胸痛寻求医疗帮助。一位护士使用人工智能程序来诊断原因。为什么这个系统可能实际上并不真正智能？

• 如果患者提供了完整的医疗历史，程序才能表现出智能。
• 该程序只是将她的症状与专家创建的系统中的步骤相匹配。
• 如果患者之前来过，程序才算是智能的。

第 94 题

How does an artificial neural network learn?

• A computer scientist programs each neuron to have the correct answer to any question.
• Only correct answers go into the input layer, so it learns what’s correct from the output layer.
• The hidden layers hide the incorrect answers from the rest of the network.
• It looks at the data and makes guesses, then it compares those guesses to the correct answer.

人工神经网络是如何学习的？

• 计算机科学家程序每个神经元有对任何问题的正确答案。
• 只有正确的答案进入输入层，所以它从输出层学习什么是正确的。
• 隐藏层将不正确的答案从网络的其余部分隐藏起来。
• 它观察数据并做出猜测，然后将这些猜测与正确答案进行比较。

第 95 题

The healthcare and medical insurance industries caution against using machine learning to search for patterns in data, and they do not want machines making decisions about a person’s health. Why?

• They may be decisions that humans cannot understand.
• They may be decisions that humans are unable to make.
• They may be decisions that will supplant office visits.

医疗保健和医疗保险行业警告不要使用机器学习在数据中寻找模式，并且他们不希望机器做出关于个人健康的决定。为什么？

• 这可能是人类无法理解的决定。
• 这可能是人类无法做出的决定。
• 这可能是将取代面诊的决定。

第 96 题

What type of impact does artificial intelligence have on robotics?

• AI systems can create robots that can more easily learn new tasks.
• AI systems will help robots do very precise work.
• Newer robots won’t be able to do anything without artificial intelligence.
• AI systems can help robots do simple repetitive work.

人工智能对机器人学有什么类型的影响？

• AI 系统可以创建能够更容易学习新任务的机器人。
• AI 系统将帮助机器人进行非常精确的工作。
• 新型机器人如果没有人工智能将无法做任何事情。
• AI 系统可以帮助机器人做简单重复的工作。

第 97 题

What impact will the Internet of Things (IoT) have on artificial intelligence?

• IoT devices will keep artificial intelligence agents from becoming strong AI.
• These devices can act as experts to help program expert systems.
• These devices will be a great new source of “real world” data.
• IoT devices can form neurons to help create artificial neural networks.

物联网 (IoT) 将对人工智能有什么影响？

• IoT 设备将阻止人工智能代理成为强 AI。
• 这些设备可以作为专家帮助编程专家系统。
• 这些设备将是“现实世界”数据的一个极好的新来源。
• IoT 设备可以形成神经元帮助创建人工神经网络。

第 98 题

A new online camping goods store wants to find connections between products customers buy and other products they might buy. Why would the company use unsupervised learning?

• Supervised learning is unable to identify connections between unrelated products.
• Connections can be found with any input required by the user.
• It does not yet have enough customers to make supervised learning meaningful.

一个新的在线露营用品商店想要找出顾客购买的产品和他们可能会购买的其他产品之间的联系。公司为什么会使用无监督学习？

• 监督学习无法识别不相关产品之间的联系。
• 与用户所需的任何输入都可以找到联系。
• 它还没有足够的顾客使监督学习有意义。

第 99 题

You’re a preschool worker and you want to teach your class the letters in the alphabet. So you draw the letter “B” on the board. Then you ask the two-year-old students to find a block with that same letter. Some of the students correctly find the blocks with the letter “B”, but some of the students confuse the letter “B” with the letter “D”. So the incorrect students compare their block to the letter “B” on the board, recognize the error and then decide to get another block. What type of learning is this?

• machine learning
• reinforcement learning
• supervised learning
• unsupervised learning

你是一名幼儿园工作人员，你想教你的班级字母表中的字母。所以你在黑板上画了一个“B”字母。然后你让两岁的学生找到一个带有相同字母的积木。一些学生正确地找到了带有字母“B”的积木，但有些学生将字母“B”与字母“D”混淆了。因此，错误的学生将他们的积木与黑板上的字母“B”进行比较，认识到错误，然后决定去拿另一个积木。这是什么类型的学习？

• 机器学习
• 强化学习
• 监督学习
• 无监督学习

第 100 题

Why might you want to use reinforcement learning instead of unsupervised learning?

• Reinforcement learning doesn’t require training and test data in the same way as unsupervised learning.
• Reinforcement learning allows the machine to make predictions and create strategies instead of just clustering the data.
• Reinforcement learning is a great way to cluster data based on items that are frequently bought together.
• Reinforcement learning allows the machine to create binary classifications based on labeled data.

为什么你可能想使用强化学习而不是无监督学习？

• 强化学习不像无监督学习那样需要训练和测试数据。
• 强化学习允许机器做出预测和创建策略，而不仅仅是对数据进行聚类。
• 强化学习是根据经常一起购买的项目对数据进行聚类的好方法。
• 强化学习允许机器根据标记数据创建二元分类。

第 101 题

What is one of the greatest challenges with supervised learning binary classification?

• You need a lot of pre-classified or labeled data for the training set.
• These systems are complex and inherently unreliable.
• You have to come up with multiple classifications.
• You have to let the machine come up with its own classification labels.

监督学习二元分类最大的挑战之一是什么？

• 你需要大量预分类或标记的数据作为训练集。
• 这些系统复杂且本质上不可靠。
• 你必须提出多个分类。
• 你必须让机器自己提出分类标签。

第 102 题

You work for a company that’s selling electric cars to consumers. The company wants to get the maximum amount of value from its advertising dollars. So it wants to ramp up advertising when it thinks that customers would be most interested in purchasing an electric car. Your data science team wants to create a regression analysis based on fuel prices. How might this look on an XY diagram?

• Put electric car sales on both the X and Y axis.
• You don’t need to put anything on the Y axis; just put fuel prices on the X axis and the trendline.
• Create a trendline with fuel prices along the X axis and electric car sales on the Y axis.
• Put fuel prices in dollars on the X axis and fuel prices in rupees on the Y axis.

你为一家向消费者销售电动汽车的公司工作。该公司希望从其广告费用中获得最大价值。因此，当认为顾客最有可能对购买电动车感兴趣时，它希望加大广告力度。你的数据科学团队想基于燃料价格创建回归分析。这在 XY 图上可能是什么样子？

• 在 X 轴和 Y 轴上都放电动汽车销量。
• 你不需要在 Y 轴上放任何东西；只需在 X 轴上放燃料价格和趋势线。
• 创建一个趋势线，沿 X 轴是燃料价格，Y 轴是电动汽车销量。
• 在 X 轴上放以美元为单位的燃料价格，在 Y 轴上放以卢比为单位的燃料价格。

第 103 题

How is K Nearest Neighbor like the old saying, “birds of a feather flock together?”

• Multiclass classification is like a flock of birds that needs to be classified.
• Classify unknown data against the closest data that you do know.
• You want to fly through the data as quickly as possible.
• Make sure you know everything about the data before you try to classify.

K 最近邻（K Nearest Neighbor）算法如何体现了老话“物以类聚，人以群分”？

• 多类分类就像需要被分类的一群鸟。
• 将未知数据与你所知道的最近的数据进行分类。
• 你希望尽可能快地通过数据。
• 在尝试分类之前，确保你对数据了如指掌。

第 104 题

What is ensemble modeling?

• This is when you use a mix of different machine learning algorithms or data to improve the outcome.
• This is when you use machine learning to perform music composition.
• This is when you mix the training data with the test data to improve the machine learning algorithm.
• This is when you use supervised and unsupervised machine learning together to make better predictions.

什么是集成建模？

• 当你使用不同的机器学习算法或数据混合以改善结果时，这就是集成建模。
• 当你使用机器学习进行音乐作曲时。
• 当你将训练数据与测试数据混合以改进机器学习算法时。
• 当你将监督学习和无监督学习一起使用以做出更好的预测时。

第 105 题

You work for a credit card company that’s trying to do a better job identifying fraudulent transactions. So your team uses unsupervised learning to create clusters of transactions that are likely to be fraudulent. The machine identified that when customers are buying electronics, it’s much more likely to be a fraudulent transaction. So you use this model for your new fraud detection system. Then customers started to complain that they couldn’t use their credit cards to purchase any electronics. What is the challenge with your model?

• You used too much data to train the algorithm how to make predictions.
• You used unsupervised learning when you should have used supervised learning.
• You overfit the model to the data, the added complexity made it difficult to manage the system.
• You underfit the model to the data, the simple rule made too many inaccurate predictions.

你在一家信用卡公司工作，该公司试图更好地识别欺诈交易。因此，你的团队使用无监督学习创建可能是欺诈的交易群。机器识别出，当客户购买电子产品时，这种交易很可能是欺诈性的。所以你使用这个模型作为新的欺诈检测系统。然后，客户开始抱怨他们无法使用信用卡购买任何电子产品。你的模型存在什么挑战？

• 你使用了太多数据来训练算法如何进行预测。
• 你在应该使用有监督学习的情况下使用了无监督学习。
• 你让模型过度拟合数据，增加的复杂性使得系统难以管理。
• 你让模型欠拟合数据，简单的规则产生了太多不准确的预测。

第 106 题

How does the bias-variance trade-off affect machine learning?

• If the machine makes a change to one, it must consider how the other is affected.
• The machine will adjust both until there is low bias and low variance.
• The machine will get either bias or variance low, which will then bring the other to low.

偏差-方差权衡如何影响机器学习？

• 如果机器对其中一个进行了更改，它必须考虑另一个如何受到影响。
• 机器将调整两者，直到偏差和方差都低。
• 机器将使偏差或方差之一降低，这将随后使另一个降低。

第 107 题

Kira is building a neural network to identify customer returns using binary classifications of defective or unsatisfied. In which layer of this neural network will Kira have a probability score?

• the hidden layers
• the input layer
• the output layer

Kira正在构建一个神经网络，使用二元分类来识别客户退货，分类为有缺陷或不满意。在这个神经网络的哪一层，Kira会得到一个概率分数？

• 隐藏层
• 输入层
• 输出层

第 108 题

You work for a security firm that wants to use an artificial neural network to create a video facial recognition system. So you create a training set with hundreds of images of people that are found in your video footage. You initialize the artificial neural network with random weights assigned to all its connections. When you feed through the first few images, the system does a terrible job identifying whether those people are included in the video. What would the artificial neural network now do to try and improve?

• It will reinitialize and add random weights to all the connections.
• It will adjust the weights of the connections to see if it does a better job making a prediction.
• It will add weight to the data to do a better job identifying the image in the network.
• It will add more layers to the output layer to see if it does a better job making a prediction.

你为一家安全公司工作，该公司想要使用人工神经网络创建一个视频人脸识别系统。因此，你创建了一个训练集，包含在你的视频资料中找到的数百张人物图片。你用随机权重初始化人工神经网络的所有连接。当你输入前几张图片时，系统在识别这些人是否包含在视频中方面做得非常糟糕。人工神经网络现在会做什么尝试改进？

• 它将重新初始化并为所有连接添加随机权重。
• 它将调整连接的权重，看它是否在做出预测方面做得更好。
• 它将为数据添加权重，以更好地在网络中识别图像。
• 它将在输出层添加更多层，看它是否在做出预测方面做得更好。

第 109 题

With an artificial neural network, what is the point of having a cost function?

• It helps the network determine the cost of the error so they can make larger or smaller adjustments to its guesses.
• It helps the network determine whether there should be many more hidden layers in the network.
• It shows that the network should make the same level of adjustment whether it’s 67% right or 99% right.
• It shows that at some point the processing power cost will be too great for the neural network to make accurate predictions.

在人工神经网络中，成本函数的作用是什么？

• 它帮助网络确定错误的成本，以便它们可以对其猜测进行更大或更小的调整。
• 它帮助网络确定是否应该在网络中增加更多隐藏层。
• 它表明，无论网络正确率是67%还是99%，网络应该做出相同水平的调整。
• 它表明，在某些点上，处理能力的成本将对于神经网络做出准确预测来说过于巨大。

第 110 题

How can you best describe the cost function as it applies to neural networks?

• a measure of how accurate a machine learning estimate is
• the amount of money spent to develop a neural network
• a number the system uses to measure its answer against the correct answer

你如何最好地描述成本函数在神经网络中的应用？

• 一个衡量机器学习估计准确性的度量。
• 开发一个神经网络花费的金额。
• 系统用来将其答案与正确答案进行比较的一个数字。

第 111 题

You work for an organization that created an artificial neural network to estimate home values. The data science team said that they had created 1,100 parameters for the data model to use to make predictions. Now the data model is so complex that it takes enormous processing power and still has trouble making accurate estimates. What might be the challenge with this data model?

• The model is underfitting the data.
• The model is underutilizing the data.
• The model is not generating new data.
• The model is overfitting the data.

你为一个组织工作，该组织创建了一个人工神经网络来估算房屋价值。数据科学团队表示，他们为数据模型创建了1,100个参数以用于进行预测。现在，这个数据模型如此复杂，以至于需要巨大的处理能力，而且仍然难以做出准确的估算。这个数据模型可能面临的挑战是什么？

• 模型对数据欠拟合。
• 模型未能充分利用数据。
• 模型未生成新数据。
• 模型对数据过拟合。

第 112 题

In 1980, the philosopher John Searle created an analogy called the “Chinese room argument.” He explained how one could stand inside a room, armed with a Chinese phrase book, and simulate a conversation with someone communicating via a slot on the door. What was the point of this analogy?

• It’s very easy for systems to become intelligent through language.
• It’s very easy for these computer systems to learn new languages if they have access to the data.
• Sometimes systems can seem intelligent, but they’re just mindlessly matching patterns.
• The Chinese language is very similar to computer programming, with set symbols and definitions.

1980年，哲学家约翰·塞尔（John Searle）创造了一个称为“中文房间论证”的类比。他解释了如何站在一个房间里，手持一本中文短语书，通过门上的一个小口与人进行模拟对话。这个类比的重点是什么？

• 系统通过语言变得智能是非常容易的。
• 如果这些计算机系统能够访问数据，它们学习新语言是非常容易的。
• 有时系统看似智能，但它们只是在无意识地匹配模式。
• 中文与计算机编程非常相似，都有固定的符号和定义。

第 113 题

Which device would you describe as being a part of the Internet of Things?

• A chess computer that uses machine learning to teach the system the game by playing against itself.
• An artificial neural network that uses reinforcement learning to make novel recommendations.
• A ring that records the quality of your sleep and uploads it to a website.
• A self-driving car that relies on enormous amounts of data.

您会将哪种设备描述为物联网的一部分？

• 一个使用机器学习，通过自我对弈来教系统下棋的棋类计算机。
• 一个使用强化学习做出新颖推荐的人工神经网络。
• 一个记录您睡眠质量并将其上传到网站的戒指。
• 一个依赖大量数据的自动驾驶汽车。

第 114 题

Artificial intelligence will always seem the most impressive within a world of

• Human-created content.
• Set rules and data.
• Spoken words and text.
• Audio and video recordings.

人工智能在以下哪个领域内总是显得最为令人印象深刻？

• 人类创建的内容。
• 固定的规则和数据。
• 口语词汇和文本。
• 音频和视频录制。

第 114 题

Binary classification uses which type of learning?

• Supervised
• Unsupervised
• General artificial
• Reinforcement

二元分类使用哪种类型的学习？

• 监督学习
• 无监督学习
• 通用人工智能
• 强化学习

第 115 题

What is one of the biggest differences between supervised and unsupervised machine learning?

• Supervised machine learning requires a lot of processing power, while unsupervised machine learning is very lightweight.
• Supervised machine learning works with machine learning algorithms, while unsupervised machine learning is always done by an artificial neural network.
• Supervised learning classifies data into human-created categories, while unsupervised learning clusters data into machine-created groups.
• Supervised machine learning works with massive datasets, while unsupervised learning doesn’t really use very much data.

监督式和非监督式机器学习之间最大的差异之一是什么？

• 监督式机器学习需要大量的处理能力，而非监督式机器学习非常轻量。
• 监督式机器学习使用机器学习算法，而非监督式机器学习总是由人工神经网络完成。
• 监督学习将数据分类到人类创建的类别中，而非监督学习将数据聚类到机器创建的组中。
• 监督式机器学习处理大量数据集，而非监督学习实际上并不使用很多数据。

第 116 题

You work for a company that wants to improve spam filtering for mobile email applications. Your data science team gathers one million messages that have been correctly labeled as spam. You then train an artificial neural network to correctly identify these spam messages. After you train the system, one of the product managers asks why you don’t use those same million messages to test the network for accuracy. How should you respond?

• If you use the training data, then you’re not testing how well the system will do in the future to identify spam.
• It is a good idea to use the same messages, but the machine learning system can test its accuracy.
• An artificial neural network does not use test data like other machine learning systems.
• That is an efficient way to train the system without having to find another several million email messages.

你在一家希望改善移动电子邮件应用程序的垃圾邮件过滤功能的公司工作。你的数据科学团队收集了一百万条已正确标记为垃圾邮件的信息。然后，你训练一个人工神经网络以正确识别这些垃圾邮件。在你训练系统之后，一位产品经理问为什么不使用这同样的一百万条消息来测试网络的准确性。你应该如何回应？

• 如果你使用训练数据，那么你就不能测试系统将来识别垃圾邮件的效果如何。
• 使用相同的消息是个好主意，但机器学习系统可以测试其准确性。
• 人工神经网络不像其他机器学习系统那样使用测试数据。
• 这是一种高效的训练系统的方式，无需找到另外几百万封电子邮件消息。

第 117 题

You work for a large credit card company. The company has successfully created an artificial neural network that is 98% successful in predicting when someone is going to have trouble paying their bill. When someone is identified as having trouble, the system lowers their limit and blocks large purchases. Yet with millions of customers, there is still that 2% that is misidentified. These customers call in and complain. But since the system was trained on millions of billing statements, no one is sure why the machine saw the patterns. What is the name of this challenge?

• This is the training challenge that you get with customer service representatives.
• This is the challenge of using unlabeled data with unsupervised machine learning.
• This is a “black box.” Humans cannot understand why the system makes the decisions it does.
• This is the challenge of picking the incorrect machine learning algorithm.

你在一家大型信用卡公司工作。该公司成功创建了一个人工神经网络，该网络在预测某人支付账单时会遇到困难的准确率达到98%。当有人被识别为将遇到困难时，系统会降低他们的限额并阻止大额购买。然而，在数百万客户中，仍有2%被错误识别。这些客户打电话来投诉。但由于系统是在数百万计的账单声明上训练的，没有人确切知道机器是如何看到这些模式的。这个挑战叫什么名字？

• 这是你在客户服务代表培训中遇到的挑战。
• 这是使用未标记数据进行无监督机器学习的挑战。
• 这是一个“黑盒子”挑战。人类无法理解系统做出决定的原因。
• 这是选择不正确的机器学习算法的挑战。

第 118 题

Why does an artificial neural network use backpropagation?

• It’s a way for the network to go backwards and tune the weights and bias to improve accuracy.
• It’s a way to automatically apply bias to each of the neurons to adjust for faulty data.
• It’s a way to make sure that the network does not have any errors.
• It’s a way for the network to go backwards and try a different machine learning algorithm.

为什么人工神经网络要使用反向传播？

• 这是一种让网络倒退并调整权重和偏置以提高准确性的方法。
• 这是一种自动对每个神经元应用偏置以调整错误数据的方法。
• 这是一种确保网络没有任何错误的方法。
• 这是一种让网络倒退并尝试不同机器学习算法的方法。

第 119 题

What is a good definition of Q-learning?

• It is the type of unsupervised machine learning that creates Q clusters based on similar attributes.
• It is the type of reinforcement learning that helps to improve the quality of the outcome.
• It is a type of generative artificial intelligence that creates new images from an existing catalog.
• It is a type of supervised machine learning that can learn about Q data from its nearest neighbor.

Q学习的一个好的定义是什么？

• 它是一种基于相似属性创建Q簇的无监督机器学习类型。
• 它是一种帮助提高结果质量的强化学习类型。
• 它是一种从现有目录创建新图像的生成型人工智能类型。
• 它是一种可以从其最近邻居学习Q数据的监督式机器学习类型。

第 120 题

You work in the marketing department for a large company, and you’d like to create a weekly opinion letter using ChatGPT to give your take on the top news in your industry. You create a few test posts, and you notice that ChatGPT is getting the dates wrong and is mixing up the CEOs of different companies. Why are you running into this challenge?

• ChatGPT shouldn’t be used for creative writing because it’s still prone to factual errors.
• There’s a good chance that these are human errors that can be corrected by fully embracing ChatGPT.
• ChatGPT needs to scale up so that it has a better understanding of your industry.
• ChatGPT is getting much better at opinion-based writing, so you should use it now to get ahead of the game.

你在一家大公司的营销部门工作，你想使用ChatGPT创建一份每周观点信，发表你对行业头条新闻的看法。你创建了一些测试帖子，但你注意到ChatGPT在日期上出错，且将不同公司的CEO混淆了。你为什么会遇到这个挑战？

• ChatGPT不应该用于创意写作，因为它仍然容易出现事实错误。
• 这些很可能是人为错误，通过完全接受ChatGPT可以纠正。
• ChatGPT需要扩展，以便更好地理解你的行业。
• ChatGPT在基于观点的写作上变得更好，所以你现在应该使用它，以走在游戏的前面。

第 121 题

What is one way that generative AI will change how people work?

• It will increase the number of software developers that organizations need.
• It will allow people to work for AI managers who have better communication skills.
• It will allow people to focus more on creative and strategic activities.
• It will make it much easier for people to work from home.

生成式人工智能将如何改变人们的工作方式之一？

• 它将增加组织需要的软件开发人员数量。
• 它将允许人们为拥有更好沟通技巧的AI管理者工作。
• 它将让人们更多地专注于创造性和战略性活动。
• 它将使人们更容易从家中工作。

第 122 题

Reasoning engines like ChatGPT use something called a large language model to help the system come up with a human-like response. How do these large language models generate text?

• They use probability to predict the next words in response to a given prompt.
• They rank the words in a response and then go through them in a linear fashion.
• They string together several search engines to create one massive response database.
• They crawl the web in real time to train the response to match your prompt.

像ChatGPT这样的推理引擎使用一种称为大型语言模型的东西来帮助系统产生类似人类的回应。这些大型语言模型是如何生成文本的？

• 它们使用概率来预测给定提示后的下一个词。
• 它们对回应中的词进行排名，然后以线性方式逐个处理。
• 它们将几个搜索引擎串联起来，创建一个庞大的响应数据库。
• 它们实时爬取网页，训练响应以匹配你的提示。

第 123 题

Why would you want to use a reasoning engine over a search engine for gathering information?

• Search engines are inherently conversational, so sometimes when you’re researching you just want to have access to the raw data.
• A reasoning engine will give you a ranked list of the latest information available.
• A reasoning engine is designed to understand and interpret human language, which allows it to have a conversation.
• Search engines have been around for several decades and so might not have the latest data.

为什么你会想用一个推理引擎而不是搜索引擎来收集信息？

• 搜索引擎本质上是对话式的，所以有时候当你在研究时，你只想要访问原始数据。
• 推理引擎会给你一个最新信息的排名列表。
• 推理引擎被设计用来理解和解释人类语言，这使它能够进行对话。
• 搜索引擎已经存在了几十年，所以可能没有最新的数据。

第 124 题

What does the term “prompt engineering” mean?

• It’s a term for the computer scientists that assemble the system or “prompt” the system to life.
• This is when you try to get good results from the reasoning engine with short and effective prompts.
• It’s a type of search engine that focuses on maintaining the web index database.
• It’s when an engineer prompts the system through a regular series of reboots.

“提示工程”这个术语是什么意思？

• 这是指组装系统或者“启动”系统的计算机科学家们的术语。
• 这是当你尝试用简短有效的提示从推理引擎获取好结果时。
• 这是一种专注于维护网络索引数据库的搜索引擎。
• 这是当工程师通过一系列常规的重启来提示系统时。

第 125 题

What is notable about this ChatGPT prompt? “Explain the political system of the United States by using the characters and language from the Lord of the Rings.”

• You were using an analogy to better understand the concept.
• You’re creating an impossible prompt because the Lord of the Rings is fictional.
• You were violating the copyright protection from the Lord of the Rings.
• You are asking the system to make a political judgment, which it’s not designed to do.

这个ChatGPT提示有何特别之处？”使用《指环王》的角色和语言解释美国的政治体系。”

• 你在使用一个类比来更好地理解这个概念。
• 你在创建一个不可能的提示，因为《指环王》是虚构的。
• 你在侵犯《指环王》的版权保护。
• 你在要求系统做出政治判断，而这不是它被设计来做的。

第 126 题

Why is it much riskier for reasoning engines to be wrong than search engines?

• Reasoning engines are conversational, so they can be much more convincing.
• Reasoning engines are much more modern, so the technology should be much more accurate.
• Since the reasoning engine is ranking the results, it’s difficult to keep track of the answer.
• Reasoning engines usually take a long time to answer, so you want to make sure that it’s right.

为什么对于推理引擎来说，出错的风险比搜索引擎要大得多？

• 推理引擎是对话式的，所以它们可以更有说服力。
• 推理引擎更加现代化，所以技术应该更加准确。
• 由于推理引擎正在对结果进行排名，很难跟踪答案。
• 推理引擎通常需要较长时间来回答，所以你希望确保它是正确的。

第 127 题

What does it mean for a reasoning engine to have a “hallucination?”

• The reasoning engine runs out of graphic processing units or GPU cycles.
• The system presents a confident response that is actually false.
• The reasoning engine reaches artificial general intelligence (AGI).
• The system doesn’t understand your question, so it refuses to answer.

推理引擎出现“幻觉”是什么意思？

• 推理引擎用完了图形处理单元或GPU周期。
• 系统提出了一个实际上是错误的自信回应。
• 推理引擎达到了人工通用智能（AGI）。
• 系统不理解你的问题，所以它拒绝回答。

第 128 题

How is the rise of generative AI most like the invention of photography?

• Photography generated a whole class of new occupations.
• Generative AI is an excellent way to capture digital data.
• Like photography, generative AI puts technology in the hands of everyday users.
• Like photography, people don’t have to rely on an artist to capture reality.

生成式人工智能的兴起最像摄影的发明在哪些方面？

• 摄影产生了一整类新的职业。
• 生成式AI是捕捉数字数据的绝佳方式。
• 像摄影一样，生成式AI将技术交到了普通用户手中。
• 像摄影一样，人们不必依赖艺术家来捕捉现实。

第 129 题

Your company wants to use generative AI to come up with new pharmaceuticals. This system will analyze all existing chemical compounds and try to develop new compounds based on the success of some of your current pharmaceuticals. This system will require a lot of custom programming and access to your proprietary data sets. What type of generative AI system might work best?

• Use a text to graphics engine, such as DALL-E 2.
• Develop your own generative AI model based on your existing data.
• Use a generative AI service like ChatGPT.
• Combine a series of open-source models and run on a cloud service.

你的公司想使用生成式AI来研发新的药物。这个系统将分析所有现有的化学化合物，并试图基于你们当前一些药物的成功来开发新的化合物。这个系统将需要大量的定制编程和访问你的专有数据集。哪种类型的生成式AI系统可能最有效？

• 使用文本到图形引擎，如DALL-E 2。
• 基于你现有的数据开发自己的生成式AI模型。
• 使用像ChatGPT这样的生成式AI服务。
• 结合一系列开源模型并在云服务上运行。

第 130 题

You’re trying to get better at prompt engineering, so you decided to try a new technique. You say, “Write a 500-word essay on large language models and hallucinations from the perspective of a computer science graduate student at a university.” What technique are you using here?

• You are using role-playing to get more accurate responses.
• You are taking an adversarial approach to get both sides of the story.
• You are using a compression technique by limiting the results to 500 words.
• You’re brainstorming with the system about large language models and hallucinations.

你正在尝试提高提示工程的技能，所以你决定尝试一种新技术。你说：“从一所大学的计算机科学研究生的角度写一篇关于大型语言模型和幻觉的500字论文。”你在这里使用的是什么技术？

• 你正在使用角色扮演来获得更准确的响应。
• 你正在采取对抗性方法来获取故事的两面。
• 你正在通过限制结果为500字来使用压缩技术。
• 你正在与系统就大型语言模型和幻觉进行头脑风暴。

第 131 题

How does a traditional web search engine work?

• Have generative AI crawl a web index.
• By indexing and using generative AI for ranking.
• By crawling, indexing, and ranking web pages.
• By searching with large language models.

传统的网络搜索引擎是如何工作的？

• 让生成式AI爬取网页索引。
• 通过索引并使用生成式AI进行排名。
• 通过爬取、索引和排名网页。
• 通过使用大型语言模型进行搜索。

第 132 题

Three large companies run their own generative AI chatbots. All three use the chatbots as a service on their websites. Company A has released the chatbot source code and data. The other two companies have refused. Company A has said that “customers should have insight into how these systems are working.” What is this company arguing in favor of?

• Protecting international human rights by allowing these systems to freely operate.
• Keeping this technology out of the hands of bad actors.
• Greater transparency into how these systems are making decisions.
• Lowering the barrier of entry for anyone to spin up their own chatbot.

三家大公司运营着自己的生成式AI聊天机器人。所有这三家公司都在其网站上作为服务使用聊天机器人。公司A已经发布了聊天机器人的源代码和数据。其他两家公司拒绝了。公司A表示：“客户应该对这些系统的工作方式有所了解。”这家公司主张支持什么？

• 通过允许这些系统自由运作来保护国际人权。
• 将这项技术置于坏人手中。
• 增加这些系统决策方式的透明度。
• 降低任何人启动自己的聊天机器人的门槛。

第 133 题

To improve your fitness, you bought a new smartwatch. Your smartwatch uploads your data to a website for free fitness reports. After you purchased your watch, you got a letter from your health insurance company that your insurance rates will increase. The insurance company has arrived at this decision “based on new data.” You called the insurance company, and they would not disclose where they purchased your new health data, but you suspect it’s from the smartwatch. What ethical issue might your smartwatch company have violated?

• They were not transparent about how they use or potentially sell your data.
• They missed an opportunity to charge you for accessing the data on your watch.
• They did not consider that you are trying to use the fitness watch to get in shape.
• They shouldn’t sell your data without giving you a percentage of the proceeds.

为了提高你的健康水平，你买了一款新的智能手表。你的智能手表将你的数据上传到一个网站，以获得免费的健身报告。在你购买手表之后，你收到了你的健康保险公司的一封信，信中说你的保险费将会增加。保险公司表示这一决定是“基于新数据”。你打电话给保险公司，他们不会透露他们从哪里购买了你的新健康数据，但你怀疑这是来自智能手表。你的智能手表公司可能违反了哪个伦理问题？

• 他们对于如何使用或可能销售你的数据不透明。
• 他们错过了向你收费以访问手表数据的机会。
• 他们没有考虑到你是试图使用健身手表来保持身材。
• 他们不应该在不给你一定比例收益的情况下销售你的数据。

第 134 题

You work for a small healthcare company. You are creating an online chat nurse that can answer medical questions for millions of customers 24/7. The project’s engineer says that you should develop the nurse as an expert system with preprogrammed responses and not use machine learning. With ML, there is too great a risk that the nurse will give the wrong answer or “hallucinate.” Do you think the engineer has a good point?

• No, it’s actually extremely rare for generative AI systems to hallucinate, so the risks are worth the cost.
• Yes, but your healthcare company should still deploy the system in the hopes of getting greater market share.
• Yes, the nurse shouldn’t give out any advice that it hasn’t been explicitly programmed to deliver.
• No, all generative AI systems hallucinate, so users take that into account when getting a response.

你在一家小型医疗保健公司工作。你正在创建一个在线聊天护士，能够24/7为数百万客户回答医疗问题。项目工程师说，你应该将这个护士开发为一个具有预编程响应的专家系统，而不是使用机器学习。使用ML，护士给出错误答案或“幻觉”的风险太大。你认为工程师的观点有道理吗？

• 不，生成式AI系统发生幻觉的情况实际上极其罕见，所以风险是值得承担的。
• 是的，但是你的医疗保健公司仍应部署该系统，希望能获得更大的市场份额。
• 是的，护士不应该提供任何未经明确编程的建议。
• 不，所有生成式AI系统都会产生幻觉，所以用户在获取响应时会考虑到这一点。

第 135 题

You work for a large website that sells royalty-free images. Your executives want to create a system that uses existing royalty-free images to create a new generative AI system that produces images without paying artist royalties. At a meeting, the manager asks if it’s ethical to use the artist’s own work to eliminate their livelihood. What is a key ethical question here?

• Should the site serve the artists or the customers purchasing art?
• How can you generate the greatest revenue from the service?
• Is there a way that the system could generate art without relying on artists?
• Can the system use the art within the current licensing agreement?

你为一个大型网站工作，该网站销售免版税图片。你的执行官想要创建一个系统，使用现有的免版税图片来创建一个新的生成性AI系统，该系统能够生产图片而不用支付艺术家版税。在一个会议上，经理问使用艺术家自己的作品来消除他们的生计是否道德。这里有一个关键的道德问题是什么？

• 网站应该服务于艺术家还是购买艺术的顾客？
• 如何从该服务中产生最大的收益？
• 系统有没有一种方式可以不依赖艺术家产生艺术？
• 系统能在当前的许可协议内使用艺术作品吗？

第 136 题

You are a board member for a large social networking company. This company has been in the news a few times for ethical lapses. As a result, you ask an AI ethics subcommittee of your board to investigate the executives to see if the executives are following any ethical AI guidelines. The subcommittee reports back that the executives have an extremely relaxed view of the role of ethics in implementing AI. What first step would you want to take?

• Ensure that policies and procedures exist for ethical concerns.
• Have the board create a policies and procedures document for the executives to follow.
• Encourage the executives to have a much less relaxed view of ethical issues.
• Congratulate the executives on their focus on profitability.

你是一家大型社交网络公司的董事会成员。这家公司因几次道德失误而登上新闻头条。因此，你要求你的董事会下属的一个AI伦理小组委员会调查执行官们是否遵循任何道德AI指南。小组委员会回报说，执行官们对实施AI的道德角色有一种极其放松的看法。你会想采取什么第一步？

• 确保存在针对道德关切的政策和程序。
• 让董事会为执行官们跟随创建一份政策和程序文件。
• 鼓励执行官们对道德问题有一个更加严肃的看法。
• 祝贺执行官们专注于盈利。

第 137 题

You and a friend purchase smartphones at the same time. After you buy the phones, you both decide to apply for a credit card from the same company. Both of you are approved for the credit card, but your interest rate is 18%, while your friend’s interest rate is just 8%. You both have the same credit score and similar incomes, so you can’t figure out why it’s more expensive for you to use your card. What ethical issue might be at play here?

• The system should generate an end-user license, so users accept the higher interest rate.
• The company should always assign the highest interest rate to generate more profits.
• The system obviously needs more data, so it doesn’t randomly assign interest rates.
• There isn’t much transparency in how the credit card company assigns interest rates.

你和一个朋友同时购买了智能手机。购买手机后，你们俩决定向同一家公司申请一张信用卡。你们都获得了信用卡的批准，但你的利率是18%，而你朋友的利率仅为8%。你们俩的信用评分相同，收入也相似，所以你无法弄清楚为什么使用你的卡更贵。这里可能存在什么道德问题？

• 系统应该生成一个最终用户许可协议，因此用户接受更高的利率。
• 公司应该总是指定最高的利率以产生更多利润。
• 显然，系统需要更多数据，这样它就不会随机分配利率。
• 信用卡公司分配利率的方式缺乏透明度。

案例分析

通过四个场景的分析，我们可以看出如何将题库中的知识点与实际应用相结合，深化理解。

• 医院放射科：介绍了利用变分自动编码器（VAE）进行异常检测的案例，强调了生成模型在实际应用中的重要性。
• ChatGPT错误分析：说明了即使是高级AI模型也会有局限，强化了对模型训练和数据更新的理解。
• 药物生成AI：展示了AI在新药研发中的应用，加深了对生成模型的认识。
• 植物识别应用：提供了一个监督学习应用实例，有助于理解机器学习分类任务。

最后

利用题库备考是一条高效之路，但深入学习和实际应用才是知识的终极归宿。希望每位学习者都能在享受知识带来的乐趣中，实现自我超越，顺利通过考试，达成自己的学习和职业目标。

阅读全文