This week I came across several articles that challenge the development and utilization of AI-based system across several domains.

Ť他的一周，我遇到了几个文章，跨多个域挑战基于人工智能系统的开发和利用。

I’ve never had to genuinely reflect on the philosophical and legal aspects of my contributions as a machine learning practitioner, but this has changed after reading some interesting articles that present the consequences of AI advancement that are happening now, and those that are yet to happen.

作为机器学习从业者，我从来不需要真正地思考过我所做贡献的哲学和法律方面，但是在阅读了一些有趣的文章之后，情况发生了变化，这些文章介绍了现在正在发生的人工智能发展的后果，以及尚未出现的事情。发生。

Our lives today could look entirely different tomorrow.

今天我们的生活明天可能看起来完全不同。

本周，我发现有趣的文章涵盖了以下主题： (This week the articles I found interesting covered the following topics:)

• Why Philosophers will be the last ones standing

  为什么哲学家将是最后一站

• Facial recognition in the legal spotlight

  法律关注的面部识别

• How to develop an effective data science portfolio

  如何开发有效的数据科学产品组合

• A battle between object detection algorithms

  目标检测算法之间的斗争

面部识别巨人拒绝分享Dave Gershgorn的算法数据集详细信息 (A Facial Recognition Giant Refuses to Share Details About Its Algorithm Dataset by Dave Gershgorn)

Would you let a machine learning model that has a failure rate of 98% and a false positive rate of 81% into production?

您是否会将具有98％的故障率和81％的误报率的机器学习模型投入生产？

Well, these claimed performance figures are from a facial recognition system that is in use by the policing force in South Wales and other parts of the United Kingdom.

好吧，这些声称的性能数据来自南威尔士和英国其他地区的治安部队使用的面部识别系统。

Dave Gershgorn article starts with a description akin to the setting of a dystopian future where an overseeing governing system monitors everyone; which is hysterically a foreshadowing of a foreseeable future.

Dave Gershgorn的文章首先描述了反乌托邦的未来，在这种情况下，监督管理系统会监控每个人。 歇斯底里地预示着可预见的未来。

South Wales Police have been using facial recognition systems since 2017 and have done this in no secrecy from the public. They’ve made arrests as a result of the facial recognition system.

自2017年以来，南威尔士州警察一直在使用面部识别系统，并且这在公众场合都不是秘密的。 他们由于面部识别系统而被捕。

On the surface value the utilization of the technology to combat crime and aiding effective policing doesn’t sound any alarms, but the accuracy metrics and performance audit results of these systems that are used to conduct arrests paint a more startling picture.

从表面上看，利用该技术打击犯罪和协助有效的治安并没有发出任何警报，但这些用于逮捕的系统的准确性指标和绩效审核结果却描绘出更加惊人的画面。

Dave’s article includes information of a lawsuit that has been taken against the governing bodies using the facial recognition software. The lawsuit is due to the facial recognition system inherent algorithmic racial bias and ineffectiveness — a subject matter that’s made headlines lately.

Dave的文章包括有关使用面部识别软件对理事机构提起诉讼的信息。 该诉讼是由于面部识别系统固有的算法种族偏见和无效性而引起的，这一主题最近成为头条新闻。

But yet, the main point of Dave’s coverage of the event is to bring to light that NEC technology(provider of the facial recognition tool) is not willing to reveal details of their dataset.

但是，Dave对事件的报道的重点是要揭示NEC技术(面部识别工具的提供者)不愿意透露其数据集的细节。

Even the police, who are using the tool has no idea of how the system is trained. In my opinion, it makes sense to expose the data that is used to train systems that have a direct/indirect effect on the public, especially in public policing.

即使是正在使用该工具的警察也不知道如何训练该系统。 我认为，公开用于培训对公众有直接/间接影响的系统的数据是有意义的，尤其是在公共警务方面。

Dave article mentions the UK influence across the rest of Europe and the world. It states that their actions and views towards facial recognition could set the precedent of how other countries in Europe move forward with the adoption and utilization of facial recognition technology.

戴夫(Dave)的文章提到了英国在整个欧洲其他国家和世界的影响力。 它指出，他们对面部识别的行为和观点可以为欧洲其他国家如何采用和利用面部识别技术开创先例。

The latter half of Dave article is brief coverage of articles that tackle complex problems within autonomous vehicle.

Dave文章的后半部分简要介绍了解决自动驾驶汽车内复杂问题的文章。

Read this article to gain an awareness of the legal consequences and actions that are taken against companies developing AI systems that are used in public settings.

阅读本文以了解法律后果以及针对开发用于公共场所的AI系统的公司所采取的行动。

这篇文章非常适合阅读： (This article is a great read for:)

• Machine learning practitioners机器学习从业者
• Technologists技术人员

忘记编码，未来的工作就是哲学作者： 卢卡·罗西 ( Luca Rossi) (Forget About Coding, The Job Of The Future Is Philosophy by Luca Rossi)

One of the most interesting article I read this week. Luca Rossi has written a piece that will send most readers down the path of self and environmental awareness.

我本周阅读的最有趣的文章之一。 卢卡·罗西(Luca Rossi )撰写了一篇文章，它将使大多数读者踏上自我意识和环保意识的道路。

After reading this article, I found my self questioning the impact of my actions and contributions that can lead to the imaginary worlds created within this article.

阅读本文后，我发现自己对自己的行为和贡献的影响会产生疑问，这些影响会导致本文产生虚构的世界。

Luca’s article starts with statements on the effect of job loss as a result of the cyclic regular occurrence of global revolutions such as agricultural, industrial and technological revolution.

Luca的文章首先阐述了由于农业，工业和技术革命等全球性革命的周期性定期发生而造成的失业问题。

He then paints a glimmer of hope with the introduction of new roles and jobs as automation renders traditional manual work obsolete. But this glimmer of hope is blocked by his personal opinions.

然后，随着自动化使传统的手工工作过时，他引入了新的角色和工作，给人一线希望。 但是，他的个人见解阻止了这种一线希望。

Luca states in his article that the non-occurrence of massive job loss due to automation and AI does not guarantee or secure a future with loads of jobs for all. Luca also expresses his concern for a future where automation reigns in all aspect of life and renders us useless; he conveys a concern for the happiness and fulfilment of man.

卢卡(Luca)在他的文章中指出，没有发生由于自动化和AI而造成的大量工作流失，并不能保证或确保所有人都有未来的工作量。 卢卡(Luca)也表达了他对未来的关注，因为未来将在生活的各个方面占据主导地位，并使我们变得无用。 他表达了对人类幸福和成就感的关注。

Luca includes a lists professions that will be eventually doomed, and as I read down the list, I was relieved not to see any machine learning roles. This relief was quickly decimated as Luca points out explicitly that positions not mentioned are still equally victim to the impending job marker apocalypse as a result of machines and automation. He even provides an example of AI emulating the creativity of man through art to support his opinions further.

Luca列出了最终将注定要失败的职业，当我阅读该清单时，我很高兴没有看到任何机器学习角色。 卢卡(Luca)明确指出，由于机器和自动化的结果，未提及的职位仍然是即将来临的工作标志启示的受害者，这种救济很快就被削弱了。 他甚至提供了一个AI通过艺术模仿人类创造力的示例，以进一步支持他的观点。

No one is truly safe. Except for Philosophers

没有人是真正安全的。 除了哲学家

Luca argues that AI can’t replace philosophers, due to the fact the philosophy is the human expression of the ambiguity in the very nature of existence and life, and this is not correlated with ‘pure intelligence’. This is an opinion that I don’t entirely agree with. I could probably get a good debate going with Luca on this subject matter.

卢卡(Luca)辩称，人工智能无法取代哲学家，这是因为哲学是存在与生命本性中模棱两可的人类表达，而这与“纯粹的智力”无关。 我并不完全同意这一观点。 我可能会和卢卡就这个主题进行一次很好的辩论。

Regardless of my opposing views on Luca opinions, I still appreciate the structured approach to his presentation of four philosophical topics that are becoming more and more relevant: morality, consciousness, the meaning of life, and alignment problem.

不管我对路卡的观点有何反对意见，我仍然赞赏他以结构化的方式介绍四个越来越重要的哲学主题：道德，意识，生活意义和结盟问题。

Luca marries each presented philosophical topics with occurrence in current technology such as self-driving cars, and hypothetical future technologies such as mind transfer and teleportation.

卢卡(Luca)将每一个提出的哲学主题都与自动驾驶汽车等当前技术以及思维转移和远距传送等假设的未来技术相结合。

By creating scenarios based on simple and complex events, Luca showcases the disparity between choices that a human could make and presents the possibility of the same decisions made by an AI system.

通过创建基于简单事件和复杂事件的场景，Luca展示了人类可以做出的选择之间的差异，并提出了AI系统做出相同决定的可能性。

Luca’s article is probably different from the more technical and straight forward articles a lot of machine learning practitioners are used to reading. But I recommend the regular consumption of articles that explores the philosophical aspects of the advancement of technology. One great book I know that does this well is Superintelligence: Paths, Dangers, Strategies by Nick Bostrom.

Luca的文章可能与许多机器学习从业者习惯阅读的技术性和直截了当的文章有所不同。 但是，我建议您定期阅读一些文章，这些文章探讨了技术进步的哲学方面。 我知道做得很好的一本好书是《 超级智能：路径，危险，策略》 (作者Nick Bostrom) 。

To conclude an interesting article, Luca boldly states that the fate of the world is in the hands of philosophers, a view I find fascinating.

在总结一篇有趣的文章时，卢卡大胆地指出，世界的命运掌握在哲学家的手中，这一观点令我着迷。

这篇文章有趣地读为： (This article is an interesting read for:)

• Technologist技术专家
• Those interested in the philosophical topics intertwined with AI那些对哲学主题感兴趣的人与AI交织在一起

如何建立有效的数据科学产品组合，作者： Harshit Tyagi (How To Build An Effective Data Science Portfolio by Harshit Tyagi)

Want a ‘stellar’ portfolio that makes you stand out from the rest of the other job candidates, then Harshit Tyagi’s article can be seen as a blueprint to building up a portfolio that achieves this goal.

想要一个“星状”的投资组合，使您在其他应聘者中脱颖而出 ，那么， Harshit Tyagi的文章可以被视为构建实现这一目标的投资组合的蓝图。

I’ll be honest and state that I did not have any portfolio to present to my current employers during the interview stage, but as Harshit very rightly states, having an advanced degree gives you an advantage. So Harshit article is well suited for Data Scientists that do not have an MSc or PhD.

我会坦白地说，在面试阶段我没有任何要向现任雇主介绍的投资组合，但是正如Harshit正确地指出的那样，拥有高级学位会给你带来好处。 因此，Harshit的文章非常适合没有MSc或PhD的数据科学家。

The first advice presented in the article is self-identification. Harshit advises job seekers to understand their current skillsets and their own limitations. Through this understanding, they will be more aware of what jobs they are best suited for.

本文中提出的第一个建议是自我识别。 Harshit建议求职者了解其当前的技能和自身的局限性。 通过这种了解，他们将更加了解自己最适合的工作。

No Data Scientist portfolio is complete without a catalogue of projects. Within the article, there are specific forms of projects readers looking to stand out from the crowd should aspire to complete. Harshit, give a near-perfect explanation of the variety of projects a job seeker can explore, and even includes several examples of portfolios.

没有项目目录，没有任何Data Scientist产品组合是完整的。 在文章中，有一些特定形式的项目，读者希望从人群中脱颖而出，并渴望完成。 Harshit对求职者可以探索的各种项目给出了近乎完美的解释，甚至还提供了一些投资组合示例。

The later steps explore the importance of online presence to assist in boosting a Data Scientist’s reputation. Platforms such as GitHub, LinkedIn, Medium and Twitter are recommended by Harshit’s, along with some personal anecdotes of how he utilizes these platform.

后面的步骤探讨了在线存在对帮助提升数据科学家的声誉的重要性。 Harshit推荐使用GitHub，LinkedIn，Medium和Twitter等平台，以及有关他如何利用这些平台的一些个人轶事。

The article concludes within some information on key items that are to be placed in a resume. Honestly, the article is riddled with great tips and information, and if you are more of a visual learner, you can also check out his YouTube channel for more great content.

本文总结了一些有关简历中关键项目的信息。 老实说，这篇文章充满了很多技巧和信息，如果您是一个视觉学习者，还可以查看他的YouTube频道以获得更多精彩内容。

本文非常适合： (This article is great for:)

• Data Science students数据科学专业的学生
• Data Science job seekers数据科学求职者

YOLOv5与Faster RCNN相比。 谁赢？ 由Priya Dwivedi (YOLOv5 compared to Faster RCNN. Who wins? by Priya Dwivedi)

If you are a very visual learner, you will enjoy the article written by Priya Dwivedi that explores the comparison of performance against two widely known object detection algorithms: YOLOv5 and Faster RCNN.

如果您是一个非常视觉化的学习者，那么您会喜欢Priya Dwivedi撰写的文章，该文章探讨了将性能与两种广为人知的对象检测算法(YOLOv5和Faster RCNN)进行比较的情况。

This is not a breakdown article of how each algorithm approach object detection, so some reader might be disappointed to find that there is no explanation of how the included algorithms works. That been said, Priya has provided a ton of links to resources that provide information on the inner workings of the algorithms.

这不是关于每种算法如何进行对象检测的分类文章，因此可能会让某些读者失望，因为他们没有说明所包含算法的工作原理。 话虽这么说，Priya提供了大量的资源链接，这些链接提供了有关算法内部工作原理的信息。

The comparison scenarios utilized by Priya to gauge the performance of each algorithm are realistic. Three scenarios are presented in this article and are in the form of videos: a video of cars, basketball match and crowded public scene.

Priya用于评估每种算法性能的比较方案是现实的。 本文以视频的形式介绍了三种情况：汽车，篮球比赛和拥挤的公共场所的视频。

In the article, Priya provides a side by side video assessment of the result of the object detection algorithms, which is then accompanied with a table of performance and accuracy criteria that each algorithm is evaluated upon.

在文章中，Priya提供了对象检测算法结果的并排视频评估，然后附有性能和准确性标准表，每种算法都在该表上进行评估。

I won’t reveal which object detection algorithm comes out on top, read the article to find out.

我不会透露哪种对象检测算法排在最前面，请阅读文章以找出答案。

The visual nature of this article makes it an easy read for deep learning practitioners of all level.

本文的视觉本质使所有级别的深度学习从业人员都可以轻松阅读。

这篇文章适合阅读： (This article is a good read for:)

• Deep learning practitioners深度学习从业人员
• Machine learning practitioners机器学习从业者

希望您觉得这篇文章有用。 (I hope you found the article useful.)

To connect with me or find more content similar to this article, do the following:

要与我联系或查找更多类似于本文的内容，请执行以下操作：

1. Subscribe to my YouTube channel for video contents coming soon here

   订阅我的YouTube频道以获取即将在这里 播出的视频内容

2. Follow me on Medium

   跟我来中

3. Connect and reach me on LinkedIn

   在LinkedIn上联系并联系我