What do the images below have in common?

以下图片有什么共同点？

Most readers will quickly catch on that they are all seats, as in places to sit. It may have taken you less than a second to recognize this common characteristic. If I heed Andrew Ng’s suggestion that anything a human can do in less than a second can be automated by a Neural Network, then I should be able to create an image classifier that recognizes seats.

大多数读者会很快流行起来，他们都是个席位 ，在地方坐。 您可能花了不到一秒钟的时间就意识到了这一共同特征。 如果我听取了吴安德(Andrew Ng)的建议 ，即人类可以在一秒钟之内完成的任何事情都可以通过神经网络实现自动化，那么我应该能够创建一个能够识别座位的图像分类器。

I could write a standard classifier using off-the-shelf python libraries. I can’t predict how good its confidence intervals will be. One thing I do know is that, regardless of the amount of data I feed it, the result will be fragile. It will break when tested on an image of a seat that deviates from what it has previously seen. I’ll have to show it a lot of images to cover all cases — it will be data-hungry.

我可以使用现成的python库编写标准分类器。 我无法预测其置信区间将有多好。 我确实知道的一件事是，无论我输入的数据量如何，结果都是脆弱的 。 当测试的座椅图像偏离先前的视野时，它将破裂。 我必须向它展示很多图像以涵盖所有情况-这将需要大量数据 。

In the recent decade since the proliferation of Neural Nets in the tech industry, a pressing concern has come to dominate the field of Machine Learning: “why are Neural Networks so brittle, so narrowly bounded, so poor at transferring what they have learned from one situation to a similar one?”

自从神经网络在技术行业中蓬勃发展以来的最近十年中， 一个紧迫的问题逐渐成为机器学习领域的一个主要问题 ：“为什么神经网络如此脆弱 ，如此狭bound，如此之差以至于无法从一个人那里转移学到的东西情况是否类似？”

This article proposes that the answer lies in the field’s foundation, in the basic assumptions we make when we use Neural Networks. It’s also why some in the field are suggesting that if we are to overcome this hurdle, the field as a whole has to reinvent itself.

本文提出了答案，这取决于我们使用神经网络时所做的基本假设，即领域的基础。 这也是为什么一些在该领域所提出的建议是 ，如果我们要克服这个障碍，该领域作为一个整体必须重塑自我 。

神经网络的认识论 (The Epistemology of a Neural Network)

In the classification example above, I chose to use a Neural Network to recognize seats, and in so doing, I made an unspoken assumption: that somewhere in the arrangement of coloured pixels, I can locate and construct the concept, seat. I might find it in the images themselves, or perhaps in the objects represented in those images. Either way, I believed that if I parsed enough images, enough data, maybe even 3D models of the objects themselves, I could eventually unearth what it is that makes each of them a seat.

在上面的分类示例中，我选择使用神经网络来识别座位 ，并且这样做时我做出了一个不言而喻的假设：在彩色像素排列的某个地方，我可以找到并构造座席概念。 我可能会在图像本身中找到它，或者在这些图像中表示的对象中找到它。 无论哪种方式，我都相信，如果我解析了足够多的图像，足够的数据，甚至是对象本身的3D模型，我最终都可以发现使它们每个人坐下的原因。

Consider the following quote, from a U of Montreal review of Representation Learning that summarizes this assumption:

请考虑以下来自蒙特利尔大学关于制图学习的评论中的引述，该引言总结了这一假设：

“An AI must fundamentally understand the world around us, and we argue that this can only be achieved if it can learn to identify and disentangle the underlying explanatory factors hidden in the observed milieu of low-level sensory data.”

“人工智能必须从根本上理解我们周围的世界，我们认为，只有当它能够学会识别和解开隐藏在观察到的低水平感官数据环境中的潜在解释因素时，才能实现这一目标。”

Every Neural Network classifier has, at its core, this assumption: that the source of truth is found in the statistical structure of the world, and consequently can be discovered by parsing the data objectively. Concepts are treated as high-level features. Human brains are even presumed to be machines that discover such objective patterns and create mental models based on them. A later quote from the same review says:

每个神经网络分类器的核心假设都是：在世界的统计结构中找到真相的来源，因此可以通过客观地解析数据来发现真相。 概念被视为高级功能。 人脑甚至被认为是发现这种客观模式并基于它们创建心理模型的机器。 同一篇评论的后来引述说：

“…this hypothesis is consistent with the idea that humans have named categories and classes because of such statistical structure (discovered by their brain and propagated by their culture)”

“……这个假设与人类之所以命名类别和类别的想法是一致的，因为这种统计结构(被大脑发现并通过文化传播)”

Unfortunately, this assumption is not only poorly supported, it is wrong. Whether or not you realize it, you and everybody else projects their own concepts onto the world they see. You are, with every concept, forcing the world into your mental mold, not the other way around.

不幸的是，这种假设不仅支持不力，而且是错误的。 无论您是否意识到它，您和其他所有人都将自己的概念投射到他们所看到的世界上。 无论采用哪种概念，您都在迫使世界进入您的思维模式，而不是反过来。

Look back at the images of the seats above. Although you hadn’t seen those images before, you immediately recognized them as seats. How? Did you dig up and decode some pattern that was hidden in the images themselves?

回头看看上面座位的图像。 尽管您以前没有看过这些图像，但是您立即将它们识别为座位 。 怎么样？ 您是否挖掘并解码了隐藏在图像本身中的某些图案？

Dig a little deeper. Why do you even have a concept of a seat? Why did you invent this idea, or learn it, give it a name, recognize it in the world? And what do all the objects in the images have in common that makes you unify them under that concept?

挖得更深一些。 为什么您甚至有座位概念？ 您为什么要发明这个想法，或者学习它，给它起个名字，在世界上认可它？ 图像中的所有对象有什么共同点，使您在该概念下将它们统一起来？

The answer should be fairly obvious: if you were ever feeling tired, and looking for somewhere to sit and rest, you would seek out one of these objects. A seat serves a purpose; it solves a problem for you.

答案应该很明显：如果您感到疲倦，并且想要在某个地方坐下休息，那么您将寻找其中一个物体。 座位有目的； 它为您解决了一个问题。

That’s strange. It seems that the definition of seat is not inherent in the objects themselves, it is based on where you would like to sit. Your interactions with the objects, not the objects themselves, define the concept for you. If any of the objects pictured did not feel like a welcome place to sit — if you felt you were too clean to sit on a dirty rock — you might not consider it a seat. It would stand out as an exception.

那很奇怪。 座位的定义似乎并不是对象本身固有的，它是基于您想坐在哪里的。 您与对象(而不是对象本身)的交互为您定义了概念。 如果上图所示的任何物体都不是一个受欢迎的坐姿-如果您觉得自己太干净了不能坐在一块肮脏的岩石上-您可能不会把它当作座位 。 作为例外，它将脱颖而出。

Or take the concept of food. If you browsed a dataset of pictures of food, it might include a picture of an insect. Whether you felt that image belonged in the dataset would depend on whether you yourself ate insects, as some people do. Or if, due to some improbable biological mutation, the entire human species should suddenly be unable to digest pineapples, then images of pineapples would cease to be classified as food, despite the fact pineapples themselves haven’t changed. Indeed, I can make food look like anything I wish, and if it becomes popular, then Google’s image classifiers must begin classifying it as food.

或采取食物的概念。 如果浏览了食物图片的数据集，则其中可能包含昆虫的图片。 您是否认为图像属于数据集将取决于您自己是否像某些人一样吃了昆虫。 或者，如果由于某种不可能的生物学突变，整个人类突然应该无法消化菠萝，那么尽管菠萝本身没有变化，但菠萝的图像将不再被归类为食物。 确实，我可以使食物看起来像我想要的任何东西 ，如果它变得流行，那么Google的图像分类器必须开始将其分类为食物 。

If even concepts like seat and food are so malleable and subjective, what can you say about concepts like beauty, far, or intelligent? It seems none of these are actually located in the world itself. They exist in your motivations. They can’t be found in the data. Looking for them there is a futile effort.

如果甚至座位和食物之类的概念是如此具有延展性和主观性，那么对于美 ， 远或智能等概念，您能说什么呢？ 似乎这些都不是真正位于世界本身中的。 它们存在于你的动机中。 在数据中找不到它们。 寻找他们是徒劳的。

机器中的外星人 (The Alien in The Machine)

Imagine you met an alien whose body and biology were drastically different from yours. Would you expect it to recognize a seat in the same way you do? What about food and beauty?

想象您遇到了一个外星人，其身体和生物学与您的完全不同。 您希望它以与您相同的方式识别座位吗？ 那么食物和美丽呢？

If it didn’t sit the way you sat, eat what you ate, and if it wasn’t attracted to the same types of creatures you were, it would initially fail quite badly at recognizing seat, food or beauty from a set of pictures.

如果它不像您坐着那样坐着，吃您吃的东西，并且如果它不被您所吸引的相同类型的生物所吸引，那么它最初将无法从一组图片中识别出座位 ， 食物或美丽 。 。

To compensate for this disconnect, say you train the alien to recognize food and chairs. You use flashcards, showing it one image at a time. In this way you hope to prepare it for decent human society. It’s an arduous process. Deep down, you know that there’s a good chance it will embarrass you if it finds itself in a situation that falls outside its training, and it ends up offering your guests scented candlesticks to eat.

为了弥补这种脱节，请说您训练外星人识别食物和椅子。 您使用抽认卡，一次显示一张图像。 您希望以此方式为体面的人类社会做准备。 这是一个艰巨的过程。 内心深处，您知道，如果发现自己处于超出训练范围之外的情况，很可能使您感到尴尬 ，最终您会为客人提供带有香气的烛台来食用。

How much easier would this process be if the alien ate food in the same way you did. As with a human child, almost no effort would be required to teach it the concept. It would drive its own education by its desires; indeed, satisfying its desires would itself be that education. This is the advantage of learning as a human being does.

如果外星人以与您相同的方式吃东西，那么这个过程会变得多么容易。 与人类孩子一样，几乎不需要付出任何努力就可以教给孩子这个概念。 它会根据自己的愿望来推动自己的教育； 确实，满足其愿望本身就是教育 。 这是人类学习的优势。

Every Neural Network classifier is like that alien, with no motives comparable to yours, arrived on earth and conscripted into service as a classifier. It has no legs to rest, no hunger to satiate, and no desires to quench, yet it is stuck trying to find patterns in data that it can use to recognize seat, food, and beauty. As a data scientist, you get upset when it fumbles, or bases its decisions on spurious features in the data. It seems to lack any semblance of common sense.

每个神经网络分类器都像那个外星人一样，没有任何动机可与您媲美，因此到达地球并应征入伍成为分类器。 它没有腿可以休息，也没有饥饿感可以满足，也没有想要放松的欲望，但是它试图在数据中找到可用来识别座位 ， 食物和美感的模式时却陷入困境。 作为数据科学家，当数据崩溃或将决策基于数据中的虚假特征时，您会感到不高兴。 它似乎缺乏常识的表象。

班达人山脉 (A Mountain of Bandaids)

In the face of these setbacks, you may try to boost its intelligence. You increase the number of equations it can process, the number of transformations it performs, the number of samples it memorizes, and the nuance of interpolations between them.

面对这些挫折，您可以尝试提高其智能。 您可以增加它可以处理的方程式数量，执行的转换 数量，存储的样本数量以及它们之间插值的细微差别。

You employ adversarial training to try to bring it, or rather force it, back into an acceptable range of behaviours. You use LIME to spot and correct unjustified correlations. But these are only newer, better training tools to give the network the semblance that it understands. You are adding a meta-band-aid onto a heap of band-aids.

您需要进行对抗性训练，以试图将其带入或更确切地说是使其恢复到可接受的行为范围内。 您可以使用LIME找出并纠正不合理的相关性。 但是，这些只是更新，更好的培训工具，可以使网络具有它所了解的外观 。 您正在将元创可贴添加到创可贴堆上。

All the while, the meaning of the concept is still in the trainer’s mind, in your mind. There was never anything in the data that could be used to help. The A.I. is playing a game called “Guess What My Favourite Number Is” with humans who get upset when it makes a mistake.

一直以来，该概念的含义仍在教练的脑海中，在您的脑海中。 数据中从来没有任何可以帮助的东西。 人工智能正在玩一个名为“猜我最喜欢的号码是什么”的游戏，当人们犯错时会感到沮丧。

This is why Neural Networks classifiers are fragile. They are adrift in a sea of data that is meaningless to them. For every concept, they lack a core focus, an opinion that unifies the phenomena. They can’t make sense of the data because they have no motives out of which to make sense. Look again at those last two words. They are no coincidence; you don’t “find” sense, you “make” sense.

这就是神经网络分类器脆弱的原因。 他们漂泊在对他们毫无意义的数据海中。 对于每个概念，它们都缺乏核心焦点 ，即统一现象的观点 。 他们无法理解数据，因为他们没有动机去理解 。 再看一下最后两个词。 他们不是巧合。 您没有“发现”意义，而是“有道理”。

客观真理 (Objective Truth)

Have you ever walked around an old neighbourhood from your childhood, and seen the same places you used to know, but in a different light? As a teenager I didn’t really notice that my home was across the street from a daycare. As an adult who might end up having children, I look at the same building, and I note that the condos I lived in are in a good location due to their proximity to that daycare.

您是否曾经从童年时代就在一个古老的社区中走过，并曾经以不同的眼光看过您曾经认识的相同地方？ 十几岁的时候，我并没有真正注意到我的家隔日托就在街对面。 作为一个可能最终会生孩子的成年人，我看着同一座建筑，我注意到我所居住的公寓位置优越，因为它们靠近日托中心。

My motivations shape what I see and how I see it. Concepts like “family-friendly” which I would have had no reason to consider, start to seep into my awareness. Someone could have vainly tried to teach me these concepts as a teenager, but I wouldn’t have fully grasped them until I could empathize with their driving motivation.

我的动机决定着我所看到的以及如何看待它。 我本来没有理由考虑的类似“家庭友好”的概念开始渗入我的意识。 十几岁的时候，有人会徒劳地尝试教我这些概念，但是直到我对它们的驾驶动机感到同情之前，我不会完全掌握它们。

Each new goal or motivation you adopt changes the world in which you live, as if you had experienced a miniature paradigm shift.

您采用的每个新目标或动机都会改变您的生活世界，就像您经历了微小的范式转变一样。

“What a man sees depends both upon what he looks at and also upon what his previous visual-conceptual experience has taught him to see ”

“一个人所看到的东西不仅取决于他所看的东西，还取决于他以前的视觉概念经验教给他的东西”

“…though the world does not change with a change of paradigm, the scientist afterward works in a different world.”

“……尽管世界不会随着范式的改变而改变，但科学家后来却在另一个世界里工作。”

When Thomas Kuhn wrote the statements above, he recognized that even when it came to apparently objective scientific facts, what you see is shaped by what you are looking for.

当托马斯·库恩( Thomas Kuhn) 撰写上述声明时，他认识到，即使涉及表面上客观的科学事实，您所看到的也取决于您所寻找的东西 。

The mere act of labeling an image in a dataset implies a ground truth. This is unwarranted. Every label is a choice on the part of the labeller, according either to his personal inclination or the prevailing social consensus¹.

在数据集中标记图像的简单动作就意味着一个事实 。 这是不必要的。 每个贴标者都是贴标者的选择，根据他个人的意愿或普遍的社会共识¹。

Browse the ImageNet dataset. You’ll find that the image labels are largely defined by arbitrary, and contemporary tastes. They are highly mutable. Take the concept of spatula: I have genuinely debated with friends as to what does and does not count as one. I found a label called Wedgewood. That’s a brand. It includes anything the Wedgewood company decides to produce. These are just two examples I immediately spotted browsing the utensil section.

浏览ImageNet数据集 。 您会发现图像标签主要由任意和现代的品味定义。 它们是高度易变的。 用抹刀的概念：我真的和朋友们争论过什么是什么，什么不算什么。 我找到了一个名为Wedgewood的标签。 那是一个品牌。 它包括Wedgewood公司决定生产的任何产品。 这些只是我立即在浏览器皿部分时发现的两个示例。

If even concrete concepts like those are arbitrary and susceptible to wide-scale revision as cultural tastes change, what can we conclude about “objectivity” except that it is at best a provisional consensus?

如果像这样的具体概念是任意的，并且随着文化品味的变化而易于大规模修改，那么除了“客观性”充其量只是暂时的共识外，我们还能得出什么结论？

哲学选择 (A Philosophical Choice)

In this article I’ve shown many examples that demonstrate the motivations and subjectivity at the root of all concepts. Despite all this, there are some readers who will be unable to accept the notion that concepts, like beauty, are born in the eye of the beholder. These readers can’t be convinced away from the belief that there are patterns in reality itself which underlie every concept. “The concept of food”, they’ll say, “is a part of the fabric of the universe, not something subjective to me”. They may even look at their loved ones and say “I find you beautiful, not because of my personal attachments, but as a consequence of the symmetry in your face”. Exceptions to these rules will not phase them.

在本文中，我展示了许多示例，这些示例从根本上说明了动机和主观性。 尽管如此，仍有一些读者无法接受这样的观念，即像美一样的概念是在情人眼中诞生的 。 不能相信这些读者相信现实中存在着构成每个概念基础的模式。 他们会说：“食物的概念”是宇宙结构的一部分，而不是对我主观的东西。 他们甚至可以看着亲人，说：“我发现你很美丽，不是因为我的个人依恋，而是因为你的脸对称。” 这些规则的例外情况不会分阶段进行。

I have no doubt such readers will mull over the dilemma in the image above, then brush it off as insignificant. I’d remind those readers that science only progresses when people detect and investigate exceptions to the dominant paradigm.

我毫不怀疑，这样的读者会仔细考虑上图中的两难处境，然后将其视而不见。 我要提醒那些读者，科学只有在人们发现并调查占主导地位的范式的例外时才能发展。

The most exciting phrase to hear in science, the one that heralds new discoveries, is not “Eureka!” but “That’s funny …” — Unknown

科学界听到的最令人兴奋的短语，预示着新发现，不是“尤里卡！”。 但是“真有趣……”-未知

The idea that concepts arise from within reality itself, and hence can be sought out in data, is a comfortable position to hold, regardless of the evidence to the contrary. It lets a person think that their beliefs match the truth, and are therefore beyond doubt or reproach. And since such people think they are being objective, they can’t easily be convinced otherwise².

概念是从现实本身内部产生的，因此可以在数据中找到它的想法，无论有相反的证据如何，都可以很容易地抓住。 它使一个人认为自己的信念与事实相符，因此是毋庸置疑或无可指摘的。 而且由于这些人认为他们是客观的，因此不能轻易说服他们²。

Such people, unfortunately, have to be left behind.

不幸的是，这些人必须被抛在后面。

For the rest of us, if we want neural networks to overcome this next great challenge, to make them robust, reliable, and meaningful, we have to make the agent’s motivations a fundamental factor in its calculus of concepts.

对于我们其余的人，如果我们希望神经网络克服下一个巨大的挑战，使其变得健壮，可靠和有意义，我们就必须使主体的动机成为其概念计算的基本因素。

Once you make this epistemological leap, a lot of seemingly insurmountable problems quickly vanish. The symbol-grounding problem is resolved: concepts are based around motivations, and symbols are instantiations of concepts in a specific context. A.I. no longer needs to be fed reams of labeled data to cover all possible edge cases — it can define its knowledge by itself, based on what will help it achieve its goals. It can self-correct in cases of uncertainty. It develops common sense.

一旦有了认识论上的飞跃，许多看似无法解决的问题就会Swift消失。 符号接地问题得以解决： 概念基于动机 ，符号是特定上下文中概念的实例化。 不再需要向AI提供大量标记数据来覆盖所有可能的极端情况-AI可以根据自己的知识来定义自己的知识 ，这将有助于实现其目标。 在不确定的情况下，它可以自我纠正。 它发展了常识 。

These benefits are not without their costs. We can no longer approach training models as if they were ingesting unordered, bland, uniform samples. To truly learn concepts, not like an alien but like a human, a classifier must interact with a world that is richer than what we currently provide. This will enable it to define concepts in the context of its motives. An agent must experience what tiredness is before it learns to identify a seat. It must, to a large degree, be a free agent in its own world. All this is difficult to implement, but once implemented it is easier to intuit.

这些好处并非没有代价。 我们不再能够像对待训练模型那样摄取无序，平淡，统一的样本。 为了真正地学习概念(不是像外星人而是像人类)，分类器必须与比我们当前提供的世界更丰富的世界互动。 这将使其能够根据其动机来定义概念 。 座席必须经历疲倦才能学会确定座位 。 它必须在很大程度上成为自己世界中的自由球员。 所有这些都很难实现，但是一旦实现，就更容易理解。

Nor can we avoid this challenge. There is a prevailing, naive, hope that at some point we’ll be able to look at Neural Networks from such an angle or with such a pair of glasses that the answers will become clear; concepts will emerge as epiphenomena from within the tangle of perceptrons, perhaps in the same way they emerge in our own intricate cerebrum.

我们也无法避免这一挑战。 一种普遍的，幼稚的希望是，在某个时候，我们能够从这样一个角度或戴上这样一副眼镜来看神经网络， 答案将会变得清晰。 概念将以感知现象从感知器的纠缠中出现，也许以它们在我们自己复杂的大脑中出现的相同方式出现。

We can’t hide behind the naive optimism of Neural Networks’ blackbox. We should finally let that dream go, and dedicate ourselves to the more difficult task of creating meaning, of making sense.

我们不能掩盖神经网络的黑匣子天真乐观的背后。 我们应该终于让这个梦想去，并致力于推动创建的意义， 决策意识的更艰巨的任务。

Thanks to Graham Holker for reviewing this, and whose discussion prompted many of the arguments in this article.

感谢Graham Holker对此进行了评论，并且其讨论引发了本文中的许多论点。

[1] This article is not intended merely to argue that “subjectivity” exists. That premise was never in doubt. Rather I argue that the unifying principle behind every concept is always rooted in the observer’s motivations. Nor do I suggest that reality plays no part at all in concepts, for where else would your mind derive it’s motivations from? If we liken reality to potting soil, then a motivation is the plant, and a concept is the flower.

[1]本文的目的不是仅仅为了争辩“主观性”的存在。 这个前提是毫无疑问的。 而是我认为，每个概念背后的统一原则始终植根于观察者的动机。 我也不建议现实在概念中根本不起作用，因为您的思想还会从其他地方得到它的动机？ 如果我们将现实比作盆栽土壤，那么动机就是植物，概念就是花。

[2] It’s worth noting that the statistical approach, i.e. that truth can be found in data, is still useful, and has been proven so over centuries of scientific research. During that time, scientists have defined concepts, made hypotheses about them, then tested them against reality, that is, data. This approach has only recently become a liability, when it has been applied to cognition. We are trying to conscript data into defining concepts and hypotheses themselves, a task that used to be reserved for humans. The previous assumptions therefore no longer apply.

[2]值得注意的是，统计方法(即可以在数据中找到真相)仍然有用，并且已经在数百年的科学研究中得到了证明。 在这段时间里，科学家们定义了概念，对它们进行了假设，然后将它们与现实即数据进行了对比。 当这种方法应用于认知时，它直到最近才成为一种责任。 我们正在尝试将数据征集为定义概念和假设本身，而这项任务过去是为人类保留的。 因此，先前的假设不再适用。