杰夫·霍金斯：智能的千脑理论

对于那些真正研究难题的科学家来说，你经常能听到这种说法，我对我们的工作也有同样的感受。

And that's the kind of thing you hear a lot with scientists who who really are studying a difficult problem, and I feel that way too about our work.

你有没有和沃森讨论过你正在试图解决的问题——寻找大脑的DNA？

Have you talked to or Watson about the problem you're trying to solve, of finding the DNA of the brain?

是的。

Yeah.

事实上，弗朗西斯·克里克在他生命的后期对这个问题非常感兴趣。

In fact, Francis Crick was very interested in this in the latter part of his life.

事实上，我是通过阅读他1979年写的一篇题为《思考大脑》的论文，才开始对大脑产生兴趣的。

And in fact, I got interested in brains by reading an essay he wrote in 1979 called Thinking About the Brain.

正是读了弗朗西斯·克里克的这篇论文后，我决定放弃计算机与工程领域的职业，转而成为一名神经科学家。

And that is when I decided I'm going to leave my profession of computers and engineering and become a neuroscientist, just reading that one essay from Francis Crick.

后来我有机会在晚年见到他。

I got to meet him later in life.

我曾在索尔克研究所发表演讲，他在听众席上，之后我还和他一起喝了茶。

I had I got I spoke at the Salk Institute, and he was in the audience, and then I had a tea with him afterwards.

你知道，他对一个不同的问题感兴趣。

You know, he was interested in a different problem.

他专注于意识问题。

He was focused on consciousness.

哦，那个简单的问题。

Oh, the easy problem.

对吧？

Right?

我认为那是个误导，所以我们在这方面并没有太多交集。

Well, I think it's the red herring and so we weren't really overlapping a lot there.

吉姆·沃森，他还活着，也对这个问题感兴趣。

Jim Watson, who's still alive, is also interested in this problem.

当他担任冷泉港实验室主任时，他积极推动实验室向神经科学方向发展。

When he was director of the Cold Spring Harbor Laboratories, he was really sort of behind moving in the direction of neuroscience there.

因此，他对这个领域有个人兴趣。

And so, he had a personal interest in this field.

我曾多次与他见面。

And I have met with him numerous times.

事实上，上一次见面是一年多以前，我在冷泉港实验室做了一场关于我们工作进展的报告。

And in fact, the last time was about a little bit over a year ago, I gave a talk at Cold Spring Harbor Labs about the progress we were making in our work.

那场报告非常有趣，因为他跟我说：‘如果你不是有重要的事情要讲，就不会来这里了，所以我一定会去听你的演讲。’

And it was a lot of fun because he said, well, you wouldn't be coming here unless you had something important to say, so I'm going go attend your talk.

于是他坐在了第一排。

So he sat in the very front row.

接下来是实验室的主任，布鲁斯·斯蒂尔曼。

Next was the to director of the lab, Bruce Stillman.

所以这两位都坐在礼堂的第一排，对吧？

So these guys were in the front row of this auditorium, right?

因为吉姆·沃森在场，所以礼堂里没人愿意坐第一排。

So nobody else in the auditorium wants to sit in the front row because it's Jim Watson.

还有实验室主任。

There's the director.

然后我做了演讲，之后和吉姆一起吃了晚饭。

And and I gave a talk, then I had dinner with Jim afterwards.

但有一张我同事苏比塔亚姆·哈图克的精彩照片，当时我站在台上，激动地讲解我们这个新框架的基本原理，而吉姆·沃森则坐在椅子边缘。

But it's I there's a great picture of my colleague, Subhitatyam Hattuk, where I'm up there sort of, like, screaming the basics of this new framework we have, and Jim Watson's on the edge of his chair.

他真的坐在椅子边缘，全神贯注地盯着屏幕。

He's literally on the edge of his chair, like, intently staring up at the screen.

当他发现DNA结构时，他做的第一场公开演讲就是在冷泉港实验室。

And when he discovered the structure of DNA, the first public talk he gave was at Cold Spring Harbor Labs.

所以，有一张照片。

So and there's a picture.

有一张著名的照片，显示吉姆·沃森站在白板前，用教鞭指着某个地方，手正拉着双螺旋结构。

There's a famous picture of Jim Watson standing at the whiteboard with a with a overheated thing pointing at something with pulling at the double helix at his pointer.

那张照片看起来和我的照片非常相似。

And it actually looks a lot like the picture of me.

所以这有点好笑。

So there was it's sort of funny.

我正在谈论大脑时，吉姆·沃森正盯着本特利看。

There's an area I'm talking about the brain, and there's Jim Watson staring at Bentley at it.

当然，六十年前，他当时也站在那里，指着双螺旋结构。

And, of course, there was, you know, whatever, sixty years earlier, he was standing, you know, pointing at the double helix.

这是

It's one

生物学、科学乃至所有科学中最伟大的发现之一，那就是DNA。

of the great discoveries in in all of, you know, whatever, biology, science, all science is DNA.

你的演讲中竟然有这种回响，真有趣。

Funny that there's echoes of that in your presentation.

是的。

Yeah.

从进化的时间线和历史来看，新皮层的出现是一个重大飞跃，还只是一个小小的进步？

Do you think in terms of evolutionary timeline and history, the development of the neocortex was a big leap, or is it just a small step?

所以，如果我们从地球生命诞生之初重新再来一遍，多大可能还会发展出新皮层这个机制？

So, like, if we ran the whole thing over again from the the birth of human of life on Earth, how likely would we develop the mechanism of the neocortex?

好的。

Okay.

嗯，这是两个不同的问题。

Well, those are two separate questions.

一个是它是否是一个重大飞跃，另一个是它发生的可能性有多大。

One is was it a big leap, and one was how likely it is.

明白吗？

Okay?

它们不一定相关。

They're they're not necessarily related.

也许有关联。

Maybe correlated.

是的。

Yeah.

也许有关联。

Maybe correlated.

也许不是。

Maybe not.

我们实际上没有足够的数据来对此做出判断。

We don't really have enough data to make a judgment about that.

我会说这绝对是一个巨大的飞跃。

I would say definitely it was a big leap.

我可以告诉你为什么。

And I can tell you why.

我不认为这只是一个渐进式的步骤。

I don't think it was just another incremental step.

我稍后会说明。

I'll get that in a moment.

我真的不知道这有多可能。

I don't really have any idea how likely it is.

如果我们观察进化，我们只有一个数据点，那就是地球。

If we look at evolution, we have one data point, which is Earth.

生命在数十亿年前出现在地球上。

Life formed on Earth billions of years ago.

我们并不清楚生命是外来的、本地产生的，还是被他人引入的，但它很早就在这里出现了。

Whether it was introduced here or it created here or someone introduced it, we don't really know, but it was here early.

从生命出现到演化出多细胞生物，花了非常非常长的时间。

It took a long, long time to get to multicellular life.

而从多细胞生物演化出新皮层，又花了非常非常长的时间。

And then for multicellular life, it took a long, long time to get the neocortex.

而我们仅仅拥有新皮层几万年而已。

And we've only had the neocortex for a few hundred thousand years.

所以这简直微不足道，明白吗？

So that's like nothing, okay?

那么，这有可能吗？

So is it likely?

当然，它并不是在地球上一出现就立即发生的。

Well, it certainly isn't something that happened right away on Earth.

要达到这一点需要多个步骤，因此我认为，这不太可能在其他可能存在生命的星球上瞬间发生。

And there were multiple steps to get there, so I would say it's probably not something that would happen instantaneously on other planets that might have life.

平均而言，这可能需要数十亿年的时间。

It might take several billion years on average.

这有可能吗？

Is it likely?

我不知道。

I don't know.

但你必须存活数十亿年才能知道答案。

But you'd have to survive for several billion years to find out.

大概吧。

Probably.

这是一个巨大的飞跃吗？

Is it a big leap?

是的，我认为这与其他所有进化步骤相比都是一种质的差异。

Yeah, I think it's it is a qualitative difference in all other evolutionary steps.

如果你愿意，我可以试着描述一下。

I can try to describe that if you'd like.

当然。

Sure.

从哪个方面呢？

In which way?

是的，我可以告诉你具体是怎么回事。

Yeah, I can tell you how.

差不多吧，我们先做个简要的铺垫。

Pretty much, let's start with a little preface.

人类能够做到的许多事情，并没有明显的生存优势作为先例。

Many of the things that humans are able to do do not have obvious survival advantages precedent.

比如，我们可以创作音乐。

You know, we could create music.

这真的有生存优势吗？

Is there a really survival advantage to that?

也许有，也许没有。

Maybe, maybe not.

那数学呢？

What about mathematics?

数学真的有生存优势吗？

Is there a real survival advantage to mathematics?

嗯，也许吧。

Well, maybe.

你可以勉强这么说。

You could stretch it.

你可以试着弄清楚这些问题，对吧？

You can try to figure these things out, right?

但在大多数进化历史中，一切都具有直接的生存优势。

But most of evolutionary history, everything had immediate survival advantages to it.

所以我给你讲个故事，我喜欢这个故事，它可能真也可能不真，但故事是这样的。

So I'll tell you a story, which I like, may or may not be true, but the story goes as follows.

生物自地球生命起源以来就一直在进化，不断在这类复杂性上叠加另一类复杂性。

Organisms have been evolving for since the beginning of life here on Earth, adding this sort of complexity onto that and this sort of complexity onto that.

大脑本身也是以这种方式进化而来的。

And the brain itself is evolved this way.

事实上，大脑中存在古老的部分、更古老的部分，甚至极其古老的部分，它们不断在新结构上叠加，我们持续增加新的能力。

In fact, there's old parts and older parts and older, older parts of the brain that kind of just keeps calming on new things and we keep adding capabilities.

当我们发展出新皮层时，它最初具有明确的生存优势，比如提升了视觉、听觉、触觉，以及可能的其他大脑新功能等等。

When we got to the neocortex, initially it had a very clear survival advantage in that it produced better vision and better hearing and better touch and maybe a new brain change and so on.

但我认为，进化采用了一种机制——这是我们近期的理论观点——它利用了一种很久以前为在世界中导航、确定自身位置而演化出的机制。

But what I think happens is that evolution took a mechanism and this is in our recent theories, but it took a mechanism that evolved a long time ago for navigating in the world, for knowing where you are.

这些就是大脑古老部分中的所谓网格细胞和位置细胞。

These are the so called grid cells and place cells of an old part of the brain.

它将这种构建世界地图、在地图上定位自身并导航的机制，转化成了一种简化且理想化的版本。

And it took that mechanism for building maps of the world and knowing where you are on those maps and how to navigate those maps and turns it into a sort of a slimmed down idealized version of it.

而这种理想化的版本如今可以应用于构建其他事物的地图，比如咖啡杯的地图、手机的地图、

And that idealized version could now apply to building maps of other things, maps of coffee cups, maps of phones, maps of

几乎是概念。

Concepts almost.

这些概念，是的，不只是几乎，而是完全如此。

The concepts, yes, and not just almost, exactly.

它就这样开始复制这些东西，对吧？

And And it so just started replicating this stuff, right?

它只是越来越多，越来越多。

It's just like more and more and more.

因此，我们从专门用于解决对生存至关重要的某些问题的专用神经硬件，转变为可以应用于所有问题的通用神经硬件。

So we went from being sort of dedicated purpose neural hardware to solve certain problems that are important to survival to a general purpose neural hardware that could be applied to all problems.

现在，它已经摆脱了生存的束缚。

And now it's escaped the orbit of survival.

我们现在能够将其应用于那些让我们感到愉悦、但并非明显属于生存特征的事物，而且这似乎只在人类身上大规模地发生了。

We are now able to apply it to things which we find enjoyment, you know, but aren't really clearly survival characteristics, and that it seems to only have happened in humans to the large extent.

因此，某种程度上，我们在新皮层中已经摆脱了进化压力的引力。

And so, that's what's going on where we sort of have we've sort of escaped the gravity of evolutionary pressure in some sense in the neoc ortex.

它现在在做一些非常有趣的事情，比如发现宇宙的模型，而这些模型可能对我们并没有实际帮助。

And it now does things which are really interesting, discovering models of the universe, which may not really help us.

这重要吗？

Does it matter?

知道可能存在多重宇宙，或者知道宇宙的年龄，或者各种天体是如何运作的，这些真的能帮助我们生存吗？

How does it help us surviving knowing that there might be multiverses or that there might be, you know, the age of the universe or how do, you know, various stellar things occur?

这些根本无法帮助我们生存。

It doesn't really help us survive at all.

但我们享受这个过程，这就是发生的事情。

But we enjoy it, and that's what happened.

或者至少不是以那种显而易见的方式。

Or at least not in the obvious way perhaps.

这是必需的。

It it is required.

如果你从进化角度审视整个宇宙，为了进行星际旅行并超越我们自己的太阳而生存，这是必然要求。

If if you look at the entire universe in an evolutionary way, it's required for us to do interplanetary travel and therefore survive past our own sun.

但你知道吗？

But you know what?

咱们别太

Let's not get too

是的。

Yeah.

但你知道吗，进化是在一个时间框架内进行的。

But, you know, evolution works at one time frame.

这是生存。

It's survival.

如果你从表型的生存、个体的生存角度来看，没错。

If you think of survival of the phenotype, survival of the individual Right.

没错。

Exactly.

你所说的那些，远远超出了这个范围。

It it that what you're talking about there spans well beyond that.

所以，我没有把任何有助于我的孩子在火星上更好生存的遗传特征传给他们。

So there's no genetic I'm not transferring any genetic traits to my children that are going to help them survive better on Mars.

对。

Right.

完全是不同的机制。

Totally different mechanism.

就是这样。

That's it.

那我们来谈谈这个新的想法吧，正如你提到的，我不知道你有没有一个好名字，千……

So let's get into the the new as as you've mentioned, this idea that I don't know if you have a nice name, thousand We

称之为‘千脑智能理论’。

call it the Thousand Brain Theory of Intelligence.

我喜欢这个说法。

I like it.

你能谈谈关于概念的空间视角这类想法吗？

So can you talk about this idea of spatial view of concepts and so on?

是的

Yeah.

所以，我能不能先描述一下这个核心发现，所有的东西都源于此？

So can I just describe sort of the there's an underlying core discovery, which then everything comes from that?

这非常简单，这就是实际发生的情况。

That's a very simple this is really what happened.

我们当时深入研究如何构建对世界事物的模型，以及如何对事物进行预测。

We were deep into problems about understanding how we build models of stuff in the world and how we predictions about things.

我手里正拿着一个咖啡杯，就像这样，我的手指触碰着杯身，我的食指，然后我把它移到杯口，准备感受杯沿。

And I was holding a coffee cup just like this in my hand and my finger was touching the side, my index finger, and then I moved it to the top and I was going to feel the rim at the top of the cup.

我问了自己一个非常简单的问题。

And I asked myself a very simple question.

我说，首先，假设我的大脑在触碰之前就已经预测了会感受到什么。

I said, well, first of all, let's say I know that my brain predicts what it's going to feel before it touches it.

你可以只是想想，想象一下。

You can just think about it and imagine it.

因此我们知道，大脑一直在进行预测。

And so we know that the brain's making predictions all the time.

那么问题来了，要做出这样的预测需要什么？

So the question is, what does it take to predict that?

对吧？

Right?

这里有一个非常有趣的答案。

And there's a very interesting answer.

首先，大脑必须知道它正在触摸一个咖啡杯。

First of all, says the brain has to know it's touching a coffee cup.

它必须拥有一个关于咖啡杯的模型。

It has to have a model of a coffee cup.

它需要知道手指当前在杯子上的位置相对于杯子的关系，因为当我移动手指时，它需要知道移动完成后手指相对于杯子的位置，然后才能预测将感受到什么。

It needs to know where the finger currently is on the cup relative to the cup because when I make a movement it needs to know where it's going to be on the cup after the movement is completed relative to the cup and then it can make a prediction about what it's going to sense.

因此，这让我意识到，进行这种预测的新皮层必须知道它正在感知、正在触摸一个杯子，并且需要以杯子为参考系来知道我手指的位置。

So, this told me that the neocortex, which is making this prediction, needs to know that it's sensing, it's touching a cup and it needs to know the location of my finger relative to that cup in a reference frame of the cup.

杯子相对于我身体的位置并不重要，它的朝向也不重要，这些都无关紧要。

It doesn't matter where the cup is relative to my body, it doesn't matter its orientation, none of that matters.

重要的是我的手指相对于杯子的位置，这表明新皮层有一个以杯子为锚点的参考系，否则我无法确定位置，也无法预测我新的位置。

It's where my finger is relative to the cup, which tells me then that the neocortex has a reference frame that's anchored to the cup because otherwise I wouldn't be able to say the location and I wouldn't be able to predict my new location.

然后我们很快就能立刻想到，我皮肤的每一个部分都可能接触到这个杯子，因此我皮肤的每一个部分都在进行预测，而每一个部分都必须拥有一个用于预测的参考系。

And then we quickly, very instantly you can say, well, every part of my skin could touch this cup and therefore every part of my skin is making predictions and every part of my skin must have a reference frame that it's using to make predictions.

所以，核心观点是，在整个新皮层中，所有信息都是以参考系的形式存储和引用的。

So, the big idea is that throughout the neocortex, everything is being stored and referenced in reference frames.

你可以把它们想象成XYZ坐标系，但它们并不是那样的。

You can think of them like XYZ reference frames, but they're not like that.

我们对这背后的神经机制了解很多，但大脑是用参考系来思考的。

We know a lot about the neural mechanisms for this but the brain thinks in reference frames.

作为一名工程师，如果你是工程师，这并不令人惊讶。

And as an engineer, if you're an engineer, this is not surprising.

如果你想要构建一个咖啡杯的CAD模型，你会在某个CAD软件中打开它，然后指定一个参考系，说这个特征位于这个位置，等等。

You'd say if I wanted to build a CAD model of the coffee cup, well, I would bring it up in some CAD software and I would assign some reference frame and say this feature's at this location and so on.

但事实上，这种想法在整个新皮层无处不在，这曾经是一个新颖的观点。

But the fact that this, the idea that this is occurring throughout the neocortex everywhere, it was a novel idea.

从那以后，无数事情都迎刃而解了，真的非常多。

And then a zillion things fell into place after that, a zillion.

所以我们现在对新皮层处理信息的方式有了完全不同的理解。

So now we think about the neocortex as processing information quite differently than we used to do it.

过去，我们认为新皮层是处理感觉数据，从这些感觉数据中提取特征，再从特征中进一步提取特征，就像今天的深度学习网络那样。

We used to think about the neocortex as processing sensory data and extracting features from that sensory data and then extracting features from the features very much like a deep learning network does today.

但大脑根本不是这样工作的。

But that's not how the brain works at all.

大脑的工作方式是将一切——每一个输入、所有事物——都分配到参考框架中。

The brain works by assigning everything, every input, everything to reference frames.

在你的新皮层中，同时活跃着成千上万甚至更多的参考框架。

And there are thousands, hundreds and thousands of them active at once in your neocortex.

这听起来令人惊讶，但一旦你真正理解了这一点，就会明白它几乎解释了我们关于这一结构的所有谜团。

It's a surprising thing to think about, but once you sort of internalize this, you understand that it explains almost all the mysteries we've had about this structure.

因此，这带来的一个后果是，新皮层的每一个小部分，比如一平方毫米，这样的区域大约有15万个。

So, one of the consequences of that is that every small part of the neocortex, say a millimeter square, and there's 150,000 of those.

所以大约是15万平方毫米。

So it's about 150,000 square millimeters.

如果你取新皮层的每一平方毫米，它都会接收一些输入，并将这些输入分配到相应的参考框架中。

If you take every little square millimeter of the cortex, it's got some input coming into it and it's going to have reference frames where it's assigned that input to.

而每一平方毫米都能学会物体的完整模型。

And each square millimeter can learn complete models of objects.

那么，我这话是什么意思呢？

So, what do I mean by that?

如果我正在触摸咖啡杯，仅仅触碰一个点，我是无法了解这个咖啡杯的全貌的，因为我只感受到了一部分。

If I'm touching the coffee cup, well, if I just touch it at one place, I can't learn what this coffee cup is because I'm just feeling one part.

但假如我移动杯子，触摸它不同的部位，我就能建立起杯子的完整模型，因为我正在填充这个三维地图——也就是咖啡杯。

But if I move it around the cup and touch it at different areas, I can build up a complete model of the cup because I'm now filling in that three-dimensional map, which is the coffee cup.

我可以问自己：我在这些不同位置感受到的到底是什么？

I can say, oh, what am I feeling at all these different locations?

这就是基本理念。

That's the basic idea.

但实际情况比这更复杂。

It's more complicated than that.

因此，随着时间推移——我们之前讨论过时间——即使一个单独的皮层柱，只感知世界的一小部分，也能构建出一个完整物体的模型。

So through time, and we talked about time earlier, through time even a single column which is only looking at or a single part of the cortex which is only looking at a small part of the world can build up a complete model of an object.

所以，如果你想想大脑中接收所有手指输入的那部分区域，它们分布在头顶上方。

And so if you think about the part of the brain which is getting input from all my fingers, so they're spread across the top of your head here.

这就是体感皮层。

This is the somatosensory cortex.

每个不同的皮肤区域都对应着一个皮层柱。

There's columns associated with all the different areas of my skin.

我们相信，所有这些柱体都在构建这个杯子的模型，每一个都在构建，或者构建其他物体的模型。

And what we believe is happening is that all of them are building models of this cup, every one of them, or things.

并不是每个皮层柱或皮层的每个部分都在构建所有事物的模型，但它们都在构建某些事物的模型。

Not every column or every part of the cortex builds models of everything, but they're all building models of something.

所以当我用手触摸这个杯子时，多个关于这个杯子的模型会被激活。

And so have so when I touch this cup with my hand, there are multiple models of the cup being invoked.

如果我用眼睛看它，同样也会激活许多关于杯子的模型，因为视觉系统的每个部分——大脑并不是在处理一张完整的图像。

If I look at it with my eyes, there are again many models of the cup being invoked because each part of the visual system, the brain doesn't process an image.

这种想法具有误导性。

That's a misleading idea.

这就像你的手指触摸杯子一样，我的视网膜不同区域正在观察杯子的不同部分。

It's just like your fingers touching the cup, so different parts of my retina are looking at different parts of the cup.

成千上万个关于这个杯子的模型同时被激活。

And thousands and thousands of models of the cup are being invoked at once.

它们都在彼此投票，试图弄清楚发生了什么。

And they're all voting with each other, trying to figure out what's going on.

这就是为什么我们称之为‘千脑智能理论’，因为一个杯子并不存在单一的模型。

So that's why we call it the Thousand Brains Theory of Intelligence because there isn't one model of a cup.

这个杯子有成千上万个模型。

There are thousands of models of this cup.

你的手机、摄像头、麦克风等也有成千上万种模型。

There are thousands of models of your cell phone and about cameras and microphones and so on.

这是一个分布式的建模系统，与人们以往对它的理解非常不同。

It's a distributed modeling system, which is very different than the way people have thought about it.

这是一个非常引人入胜且有趣的想法。

And so that's a really compelling and interesting idea.

我有两个初步的问题。

I have two first questions.

首先，关于所有这些模型如何整合在一起，你有这上千个大脑。

So one, on the ensemble part of everything coming together, you have these thousand brains.

你怎么知道哪一个在形成对事物的理解上做得最好？

How how do you know which one has done the best job of forming the

好问题。

Great question.

让我试着解释一下。

Let me try to explain.

神经科学中有一个被称为传感器融合问题的问题。

There there's a problem that's known in neuroscience called the sensor fusion problem.

是的。

Yes.

所以这个想法是，图像来自眼睛。

And so the idea is something like, oh, the image comes from the eye.

视网膜上有一个图像，它被投射到新皮层。

There's a picture on the retina and it gets projected to the neocortex.

到这个时候，它已经被分散得到处都是，变得混乱而扭曲，碎片遍布各处，看起来已经不像一幅图像了。

Oh, by now it's all spat out all over the place and it's kind of squirrelly and distorted and pieces are all over the It doesn't look like a picture anymore.

那么，它们什么时候才会重新整合在一起呢？

When does it all come back together again?

对吧？

Right?

或者你可能会说，是的，但我还伴随着声音或触觉，与这个杯子有关。

Or you might say, Well, yes, but I also have sounds or touches associated with the cup.

所以我正在看这个杯子，同时也在触摸这个杯子。

So I'm seeing the cup and touching the cup.

它们是怎么再次组合在一起的呢？

How do they get combined together again?

所以这被称为传感器融合问题，就好像所有这些分散的部分都需要在某个地方整合成一个统一的模型。

So it's called the sensor fusion problem, as if all these disparate parts have to be brought together into one model someplace.

这种想法是错误的。

That's the wrong idea.

正确的想法是，所有这些部分都在投票。

The right idea is that you got all these guys voting.

存在关于杯子的听觉模型。

There's auditory models of the cup.

存在关于杯子的视觉模型。

There's visual models of the cup.

存在关于杯子的触觉模型。

There's tactile models of the cup.

在视觉系统中，有些模型可能更专注于黑白图像。

Are one in the vision system, there might be ones that are more focused on black and white.

有些则关注颜色。

One's versioning on color.

这其实并不重要。

It doesn't really matter.

有成千上万种关于这个杯子的模型。

There's just thousands and thousands of models of this cup.

它们进行投票。

And they vote.

它们实际上并不会集中到一个地方。

They don't actually come together in one spot.

你就这么想吧。

Just literally think of it this way.

想象一下，这些柱子大约像一根细意大利面那么粗。

Imagine you have these columns are like about the size of a little piece of spaghetti.

明白吗？

Okay?

大约两毫米半高，宽约一毫米。

Like two and a half millimeters tall and about a millimeter in wide.

它们并不是实体的，但你可以这样想象。

They're not physical like but you can think of them that way.

每个都在尝试猜测这是什么东西，或者在触碰它。

And each one's trying to guess what this thing is or touching.

如果允许它们移动并接触，它们能做得相当不错。

Now they can do a pretty good job if they're allowed to move over touch.

我可以把手伸进一个黑箱里，用手指在物体上移动，如果触碰到足够多的区域，我就会说：好吧，现在我知道这是什么了。

I can reach my hand into a black box and move my finger around an object and if I touch enough spaces I go, okay, now I know what it is.

但通常我们并不会那样做。

But often we don't do that.

通常，我只需伸手一把抓起某物，就能立刻认出来。

Often I can just reach and grab something with my hand all at once and I get it.

或者，如果我必须透过吸管来看这个世界，也就是说我只能看到一小列信息，我只能看到事物的一部分，因为我必须移动吸管。

Or if I had to look through the world through a straw, so I'm only invoking one little column, I can only see part of something because I have to move the straw around.

但当我睁开眼睛时，我一下子就能看到整个东西。

But if I open my eyes, I see the whole thing at once.

所以我们认为发生的情况是，所有这些像意大利面一样的小部分，也就是皮层中的所有这些小列，都在试图猜测它们所感知的是什么。

So what we think is going on is all these little pieces of spaghetti, if you will, all these little columns in the cortex are all trying to guess what it is that they're sensing.

如果它们有时间并能随着时间推移进行探索，就能做出更好的猜测。

They'll do a better guess if they have time and can move over time.

所以，如果我移动眼睛和手指。

So if I move my eyes and move my fingers.

但如果它们不能，就会做出很差的猜测。

But if they don't, they have a poor guess.

这是一种关于它们可能触碰到的东西的概率性猜测。

It's a probabilistic guess of what they might be touching.

现在想象一下，它们可以把这个概率值标注在每一小段意大利面的顶端。

Now imagine they can post their probability at the top of little piece of spaghetti.

每一个都在说：我认为它是，但这并不是一个真正的概率分布。

Each one of them says, I think it And it's not really a probability distribution.

它更像是一组可能性。

It's more like a set of possibilities.

在大脑中，它并不是以概率分布的方式运作的。

In the brain, it doesn't work as a probability distribution.

它更像我们所说的并集。

It works as more like what we call a union.

你可以这么说，一个柱状结构说：我认为它可能是咖啡杯、易拉罐或水瓶。

You could say, and one column says, I think it could be a coffee cup, a soda can, or a water bottle.

另一个柱状结构说：我认为它可能是咖啡杯，或者电话、相机之类的，对吧？

And another column says, I think it could be a coffee cup or, you know, telephone or camera or whatever, right?

所有这些都在表达它们认为它可能是什么。

And all these guys are saying what they think it might be.

在大脑的某些层中，存在着长距离的连接。

And there's these long range connections in certain layers in the cortex.

因此，某些层中的某些细胞类型会向大脑的其他区域发送投射，这就是投票发生的地方。

So there's some layers in some cell types in each column send the projections across the brain and that's the voting occurs.

因此，这是一种简单的联想记忆机制。

And so there's a simple associative memory mechanism.

我们在最近的一篇论文中描述了这一点，并且已经对此进行了建模。

We've described this in a recent paper and we've modeled this.

这意味着它们都能迅速达成一致，找到唯一或最佳的答案。

That says they can all quickly settle on the only or the one best answer for all of them.

如果存在一个最佳答案，它们都会投票并说：没错，一定是咖啡杯。

If there is a single best answer, they all vote and say, Yep, it's got to be the coffee cup.

到了那一刻，它们都明白那是一个咖啡杯。

And at that point, they all know it's a coffee cup.

到了那一刻，所有人都会像它是一个咖啡杯一样行动。

And at that point, everyone acts as if it's a coffee cup.

是的，我们知道它是个咖啡杯。

Yep, we know it's a coffee cup.

尽管我只看到过这个世界的一小部分，但我清楚我正在触摸、看到或经历的是一件咖啡杯。

Even though I've only seen one little piece of this world, I know it's a coffee cup I'm touching or I'm seeing or whatever.

因此，你可以认为这些列都在观察不同的部分、不同的位置、不同的感官输入、不同的地点，它们全都各不相同。

And so you can think of all these columns are looking at different parts, different places, different sensory input, different locations, they're all different.

但进行投票的这一层会变得稳固。

But this layer that's doing the voting, it solidifies.

它就像结晶一样，明确地说：哦，我们都明白我们在做什么。

It's just like it crystallizes and says, Oh, we all know what we're doing.

所以你并不会把这些模型整合成一个统一的模型。

And so you don't bring these models together in one model.

你只是投票，然后投票结果会结晶固化。

You just vote, and there's a crystallization of the vote.

很好。

Great.

这至少是一种令人信服的方式来理解你如何构建对世界的认知模型。

That's at least a compelling way to think about about the way you form a model of the world.

现在你提到了一个咖啡杯。

Now you you talk about a coffee cup.

据我理解，你也提出这一点，这不仅仅适用于咖啡杯，对吧？

Do you see this, as far as I understand, you are proposing this as well, that this extends to much more than coffee cups?

是的。

Yeah.

确实如此。

It does.

至少适用于物理世界。

Or at least the physical world.

它扩展到了概念的世界。

It expands to the world of concepts.

是的。

Yeah.

确实如此。

It does.

首先，支持这一观点的初步证据是，与语言、高级思维或数学等相关的大脑新皮层区域，看起来与处理视觉、听觉和触觉的新皮层区域相似。

And, well, first, the prima facie evidence for that is that the regions of the neocortex that are associated with language or high level thought or mathematics or things like that, they look like the regions of the neocortex that process vision and hearing and touch.

它们看起来没有不同，或者仅略有不同。

Don't look any different or they look only marginally different.

因此，如果提出新皮层所有部分都在执行相同功能的弗农·诺姆卡斯特是正确的，那么负责语言、数学或物理的区域也遵循同样的原理。

And so one would say, well, if Vernon Naumcastle who proposed that all the parts of the neocortex are doing the same thing, if he's right, then the parts that are doing language or mathematics or physics are working on the same principle.

它们必定是在参考框架的原理上运作的。

They must be working on the principle of reference frames.

这听起来有点奇怪。

So that's a little odd thought.

但当然，我们之前对这些机制一无所知，所以我们就先接受这个观点吧。

But of course we prior had idea how these things happen so let's go with that.

在我们最近的论文中，我们稍微讨论了一下这一点。

And in our recent paper we talked a little bit about that.

自那以后，我对此进行了更多研究。

I've been working on it more since.

我现在对它有了更好的理解。

I have better ideas about it now.

我坐在这里非常确信这就是正在发生的事情，我可以举一些例子来帮助你思考这个问题。

I'm sitting here very confident that that's what's happening and I can give you some examples to help you think about that.

我们还没有完全理解它，但我对它的理解比我在任何论文中描述的都要深入。

It's not we understand it completely but I understand it better than I've described it in any paper so far.

但我们确实提出了这个观点。

But we did put that idea out there.

它说的是，好吧，这是一个不错的起点，证据也表明这就是它运作的方式，然后我们可以逐个部分地解决这个问题。

It says, okay, this is it's a good place to start, you know, and the evidence would suggest that's how it's happening and then we can start tackling that problem one piece at a time.

那么，什么是高级思维？

Like, what does it mean to do high level thought?

什么是语言？

What does it mean to do language?

这如何融入参考框架的体系中？

How would that fit into a reference frame framework?

是的，有一个叫Anki的应用程序，能帮助你记住不同的概念，它们提到一种记忆宫殿的方法，通过在脑海中构建一个物理空间来记住完全随机的概念，是的。

Yeah, so there's a I don't know if you could tell me if there's a connection, but there's an app called Anki that helps you remember different concepts, and they they talk about, like, a memory palace that helps you remember completely random concepts by sort of trying to put them in a physical space in your mind Yeah.

把它们放在一起。

And putting them next to each other.

地点记忆法。

The method of loci.

地点。

Loci.

是的。

Yeah.

不知为什么，这种方法似乎非常有效。

For some reason, that seems to work really well.

是的。

Yeah.

这仅仅是记忆某些事实的一种非常狭窄的应用。

Now that's a very narrow kind of application of just remembering some facts

某种东西。

of something.

但这却是一个非常有说服力的例子。

But that's a very, very telling one.

好的。

Okay.

是的。

Yes.

没错。

Exactly.

所以你似乎是在描述为什么这种方法有效的原因，是的。

So it seems like you're describing a mechanism why this seems to Yes.

基本上，我们所认为的是，你所知道的一切——所有概念、所有想法、所有词语——都存储在参考框架中。

So so basically, way what we think is going on is all things you know, all concepts, all ideas, words, everything you know are stored in reference frames.

因此，如果你想记住某件事，你就必须像老鼠找到奶酪一样，或者像我的手指老鼠找到这杯咖啡一样，在参考框架中进行导航。

And so if you want to remember something, you have to basically navigate through a reference frame the same way a rat navigates to a mave and the same way my finger rat navigates to this coffee cup.

你正在某个空间中移动。

You are moving through some space.

所以，如果你被要求记住一串随机的事物，可以将它们分配到你非常熟悉的某个参照框架中，比如你的房子，对吧？

And so if you have a random list of things you're asked to remember by assigning them to a reference frame you already know very well to see your house, right?

位置记忆法的原理是，你可以说：好吧，在我的门厅里放这件东西，然后在卧室里放另一件。

And the idea of the method of locale is you can say, okay, in my lobby I'm going to put this thing and then the bedroom I put this one.

我走过走廊，把这件东西放在这里。

I go down the hall, I put this thing.

然后你想回忆这些事实。

And then you want to recall those facts.

要回忆这些事物，你只需在心理上行走，穿过你的房子。

To recall those things, you just walk mentally, you walk through your house.

你是在心理上移动于一个你早已熟悉的参照框架中。

You're mentally moving through a reference frame that you already had.

这告诉你，关于这种方法有两个非常重要的方面。

And that tells you, there's two things that are really important about it.

它告诉我们，大脑更倾向于将信息存储在参考框架中，而回忆或思考的方法就是 mentally 穿越这些参考框架。

It tells us the brain prefers to store things in reference frames and that the method of recalling things or thinking, if you will, is to move mentally through those reference frames.

你可以通过物理方式穿越某些参考框架，比如我可以实际地穿过这个咖啡杯的参考框架。

You could move physically through some reference frames, like I could physically move through the reference frame of this coffee cup.

我也可以 mentally 穿越咖啡杯的参考框架，想象自己触摸它，但我同样可以 mentally 穿越我的房子。

I can also mentally move through the reference frame of the coffee cup imagining me touching it, but I can also mentally move my house.

那么我们现在可以问自己：所有的概念都是这样存储的吗？

So now we can ask yourself, are all concepts stored this way?

最近有一些使用人类受试者进行 fMRI 的研究，我要为不知道进行这项研究的科学家姓名而道歉。

There was some recent research using human subjects in fMRI and I'm going to apologize for not knowing the name of the scientists who did this.

但他们所做的，是让人类躺在 fMRI 机器里——这是一种成像设备，并让他们完成思考鸟类的任务。

But what they did is they put humans in this fMRI machine which is one of these imaging machines, and they gave the humans tasks to think about birds.

他们使用了不同种类的鸟类，包括体型大或小、长颈或长腿的鸟等等。

So they had different types of birds and birds that looked big and small and long necks and long legs, things like that.

从 fMRI 数据中，他们得出了一项非常巧妙的实验结果。

And what they could tell from the FMRI was a very clever experiment.

你可以判断出，当人类在思考鸟类时，他们对鸟类的知识是以类似于在房间中导航时所使用的参考框架来组织的。

You get to tell when humans were thinking about the birds that the knowledge of birds was arranged in a reference frame similar to the ones that are used when you navigate in a room.

这些被称为网格细胞，当人们进行这种思考时，大脑皮层中会出现类似网格细胞的活动模式。

These are called grid cells, they are grid cell like patterns of activity in the neocortex when they do this.

这是一项非常巧妙的实验。

So it's a very clever experiment.

它基本上表明，即使你在思考抽象事物时并没有有意识地将其视为参考框架，这也告诉我们大脑实际上正在使用参考框架。

And what it basically says that even when you're thinking about something abstract and you're not really thinking about it as a reference frame, it tells us the brain is actually using frame.

而且它使用的是相同的神经机制。

And it's using the same neural mechanisms.

这些网格细胞正是我们所提出的、存在于大脑古老区域——内嗅皮层中的网格细胞的基本神经机制，而现在，类似的机制已被应用于整个大脑皮层。

These grid cells are the basic same neural mechanisms that we propose that grid cells which exist in the old part of the brain, the entorhinal cortex, that that mechanism is now, similar mechanism is used throughout the neocortex.

这种创造参考框架的有趣方式被大脑保留了下来。

It's the same nature preserve this interesting way of creating reference frames.

因此，现在他们有了实证证据，表明当你思考像鸟类这样的概念时，你正在使用基于网格细胞的参考框架。

And so now they have empirical evidence that when you think about concepts like birds that you're using reference frames that are built on grid cells.

这类似于位置法，但在这里，鸟类之间是有关联的。

So that's similar to the method of loci but in this case, the birds are related.

因此，它们创建了自己的参考框架，这个框架与鸟类空间一致。

So they create their own reference frame which is consistent with bird space.

当你思考某件事时，你会在其中穿梭。

And when you think about something, you go through that.

你可以举同样的例子。

You can make the same example.

我们以数学为例，好吗？

Let's take a mathematics, right?

假设你想证明一个猜想，好吗？

Let's say you want to prove a conjecture, okay?

什么是猜想？

What is a conjecture?

猜想是一种你认为为真但尚未证明的陈述。

A conjecture is a statement you believe to be true but you haven't proven it.

所以它可能是一个方程。

And so it might be an equation.

我想证明这个等于那个。

I want to show that this is equal to that.

在某些地方，你从那里开始。

And you some places you start with.

你说：我知道这个是对的，我知道那个也是对的，我认为要达到最终的证明，我需要经过一些中间结果。

You say, Well, I know this is true and I know this is true and I think that maybe to get to the final proof I need to go through some intermediate results.

我认为真正发生的是，这些方程或这些点被分配到了一个参考系，一个数学参考系。

What I believe is happening is literally these equations or these points are assigned to a reference frame, a mathematical reference frame.

当你进行数学运算时，一个简单的操作可能是乘法或除法，但你也可能进行变换或其他操作，这就像在数学的参考系中移动。

And when you do mathematical operations, a simple one might be multiply or divide but you might be able to transform or something else, that is like a movement in the reference frame of the math.

因此，你实际上是在尝试发现一条从数学空间中的一个位置到另一个位置的路径。

And so you're literally trying to discover a path from one location to another location in a space of mathematics.

如果你能到达这些中间结果，你就知道你的地图相当准确，也知道你使用的是正确的运算。

And if you can get to these intermediate results then you know your map is pretty good and you know you're using the right operations.

我们思考如何解决难题时，很大程度上是在为这个问题设计正确的参考框架，弄清楚如何组织信息，以及在该空间中使用哪些行为来达成目标。

Much of what we think about is solving hard problems is designing the correct reference frame for that problem, figuring out how to organize the information and what behaviors I want to use in that space to get me there.

是的。

Yeah.

所以，如果你深入探讨这种参考框架的概念——无论是数学中的参考框架，你都会从一组公理出发，试图证明猜想。

So if you dig in an an idea of this reference frame, whether it's the math, you start a set of axioms to try to get to proving the conjecture.

你能试着描述一下吗？或许退一步说，在这种背景下，你是如何理解参考框架的？

Can you try to describe, maybe taking a step back, how you think of the reference frame in that context?

是公理所适用的参考框架吗？

Is is it the reference frame that the axioms are happy in?

是包含一切的参考框架吗？

Is it the reference frame that might contain everything?

它是会变化的吗？

Is it a changing thing?

因此，你拥有许多、许多不同的参考框架。

So there it you You have many, many reference frames.

事实上，千脑智能理论认为，世界上每一件事物都有其自身的参考框架。

In fact, the way the theory, the Thousand Brains Theory of Intelligence says that every single thing in the world has its own reference frame.

所以每一个词都有其自身的参考框架，我们可以讨论这一点。

So every word has its own reference frames, and we can talk about this.

数学上是成立的，神经元完全能够做到这一点，没有任何问题。

The mathematics work out, there's no problem for neurons to do this.

但一个咖啡杯有多少个参考框架呢？

But how many reference frames does a coffee cup have?

比如，它放在桌子上。

Like, it's on a table.

假设你问的是，我手指上接触咖啡杯的那根神经元柱有多少个参考框架？

Let's say you ask how many reference frames could the column in my finger that's touching the coffee cup have?

因为咖啡杯有非常多不同的模型。

Because there are many, many models of the coffee cup.

所以并不存在一个唯一的咖啡杯模型。

So there is no one model coffee cup.

咖啡杯有多种模型，你可能会问，我的手指能学会多少种不同的东西？

There are many models of a coffee cup and you could say, Well, how many different things can my finger learn?

这就是你想问的问题吗？

Is this the question you want to ask?

想象一下，我说每一个概念、每一个想法、你所知道的任何事物，只要你能说‘我知道这个’，它都关联着一个参考框架。

Imagine I say every concept, every idea, everything you ever know about that you can say, I know that thing, has a reference frame associated with it.

当我们构建复合对象时，我们会分配参考框架来指向另一个参考框架。

And what we do when we build composite objects, we assign reference frames to point another reference frame.

所以我的咖啡杯由多个部分组成。

So my coffee cup has multiple components to it.

它有杯口、有杯身、还有把手。

It's got a rim, it's got a cylinder, it's got a handle.

这些部分各自拥有自己的参考框架，并且它们都被映射到一个主参考框架上，这个主框架就叫做‘这个杯子’。

And those things have their own reference frames and they're assigned to a master reference frame, which is called this cup.

现在，我在这上面贴上了Numenta的标志。

And now I have this Numenta logo on it.

嗯，这是世界上其他地方存在的东西。

Well, that's something that exists elsewhere in the world.

它本身就是独立的。

It's its own thing.

所以它有自己的参考框架。

So it has its own reference frame.

因此我们现在需要思考：我该如何将Numenta BOGO的参考框架赋给圆柱体或咖啡杯？

So we now have to say, well, how can I assign the Numenta BOGO reference frame onto the cylinder or onto the coffee cup?

我们在十二月发表的论文中已经讨论过这一点。

So it's all we talked about this in the paper that came out in December.

关于如何将参考框架赋给其他参考框架，以及神经元如何实现这一点的想法。

The idea of how you can assign reference frames to reference frames, how neurons could do this.

所以，我的问题是，尽管你多次提到参考框架，但我几乎觉得很有必要深入探讨一下你对参考框架的理解。

So, well, my question is, even though you mentioned reference frames a lot, I almost feel it's really useful to dig into how you think of what a reference frame is.

你提到参考框架无处不在，这对我理解它非常有帮助。

Mean, was already helpful for me to understand that you think of reference frames as something there is a lot of.

好的。

Okay.

所以我们就假设大脑中有一些神经元，其实数量并不多，大概一两万个，它们会生成大量的参考系。

So let's just say that we're going to have some neurons in the brain, not many actually, 10,000, 20,000, are going to create a whole bunch of reference frames.

这意味著什么？

What does it mean?

在这种情况下，什么是参考系？

What is a reference frame in this case?

首先，这些参考系与你可能熟悉的那些不同。

First of all, these reference frames are different than the ones you might be used to.

我们熟知很多参考系。

We know lots of reference frames.

例如，我们知道笛卡尔坐标系，也就是XYZ坐标，这是一种参考系。

For example, we know the Cartesian coordinates, XYZ, that's a type of reference frame.

我们还知道经度和纬度，这是另一种参考系。

We know longitude and latitude, that's a different type of reference frame.

如果我看着一张印刷地图，它可能有从A到M的列和从1到20的行。

If I look at a printed map, it might have columns A through M and rows one through 20.

这是一种不同类型的参考系。

That's a different type of reference frame.

它有点像笛卡尔参考系。

It's kind of a Cartesian reference frame.

大脑中参考系的有趣之处在于，我们之所以知道这一点，是因为通过研究内嗅皮层的神经科学已经证实了这一点。

The interesting thing about the reference frames in the brain, and we know this because these have been established through neuroscience studying the entorhinal cortex.

所以，我这里不是在猜测，明白吗？

So I'm not speculating here, okay?

这是大脑一个古老区域中已知的神经科学内容。

This is known neuroscience in an old part of the brain.

这些神经元创建参考系的方式是没有原点的。

The way these cells create reference frames, they have no origin.

所以，更准确地说，你有一个空间中的点，给定某个特定的移动，你就能推断出下一个点应该在哪里。

So, what is more like, you have a point, a point in some space, and given a particular movement, you can then tell what the next point should be.

然后你就能知道下一个点会是什么，依此类推。

And you can then tell what the next point would be and so on.

你可以用这种方法计算如何从一个点到达另一个点。

You can use this to calculate how to get from one point to another.

那么，我如何从我家走到我的目的地？或者我的手指如何从杯子的侧面移动到杯口？

So how do I get from my house to my home or how do I get my finger from the side of my cup to the top of the cup?

我该如何从公理推导到猜想？

How do I get from the axioms to the conjecture?

所以这是一种不同的参考框架，如果你愿意，我可以更详细地描述。

So it's a different type of reference frame, and I can if you want, I can describe in more detail.

我可以描绘一幅图景，帮助你理解这种思维方式。

I can paint a picture how you might wanna think about that.

把这想象成一种可以穿梭其中的东西，真的很有帮助。

It's really helpful to think it's something you can move through.

是的。

Yeah.

但把这看作某种意义上的空间概念有帮助吗？还是说有什么其他东西？

But is there is it is it helpful to think of it as spatial in some sense or is there something that's

没有。

No.

更准确地说，这绝对是空间的。

More It's definitely spatial.

这是一种数学意义上的空间。

It's spatial in a mathematical sense.

有多少个维度？

How many dimension?

可以是任意多的维度吗？

Can it be crazy number of dimensions?

这是一个有趣的问题。

Well, that's an interesting question.

在内嗅皮层的旧区域，他们研究了大鼠。

In the old part of the the entorhinal cortex, they studied rats.

最初看起来，这似乎只是二维的。

And initially it looks like, oh, this is just two dimensional.

就像老鼠在一个盒子或迷宫里一样。

It's like the rat is in some box in a maze or whatever.

他们使用这些二维参考框架来知道老鼠在迷宫中的位置。

And they know where the rat is using these two dimensional reference frames and know where it is in the maze.

我们说，好吧，那蝙蝠呢？

We said, okay, but what about bats?

蝙蝠也是哺乳动物，它们在三维空间中飞行。

That's a mammal and they fly in three-dimensional space.

它们是怎么做到的？

How do they do that?

它们似乎知道自己在哪里，对吧？

They seem to know where they are, right?

因此，这是一个当前活跃的研究领域，似乎内嗅皮层的神经元 somehow 能够学习三维空间。

So this is a current area of active research and it seems like somehow the neurons in the entorhinal cortex can learn three-dimensional space.

我们团队的两名成员与来自麻省理工学院的Ilha Fett刚刚在上周发表了一篇论文，论文已在bioRxiv上发布，他们展示了这些机制的工作原理——当然，如果你不想深入了解，我就不展开细节了，但网格细胞能够表征任意n维空间。

Just two members of our team along with Ilha Fett from MIT just released a paper literally last week, it's on bioRxiv, where they show that you can the way these things work, unless you want to, I won't get into the detail, but grid cells can represent any n dimensional space.

这本身并没有局限性。

It's not inherently limited.

你可以这样理解：如果是二维空间，其工作方式是叠加了许多二维切片。

You can think of it this way, if you had two dimensional, the way it works is you added a bunch of two dimensional slices.

这些机制就是这么运作的。

That's the way these things work.

存在大量二维模型，你可以用二维投影来切分任意n维空间。

There's a whole bunch of two dimensional models and you you can slice up any n dimensional space with two dimensional projections.

所以，你也可以有一个一维模型。

So, and you could have one dimensional model.

神经元实现这一机制的数学原理本身并不限制空间的维度，这一点我认为很重要。

There's nothing inherent the mathematics about the way the neurons do this which constrain the dimensionality of the space, which I think was important.

所以，显然我脑海中有一个关于这个杯子的三维地图。

So obviously I have a three-dimensional map of this cup.

也许甚至还不止这些，我不确定。

Maybe it's even more than that, I don't know.

但这显然是一幅关于这个杯子的三维地图。

But it's clearly a three-dimensional map of the cup.

我拥有的不只是杯子的投影。

I don't just have a projection of the cup.

但当我想到鸟类，或者想到数学时，也许维度会超过三维。

But when I think about birds or when I think about mathematics, perhaps it's more than three dimensions.

谁知道呢？

Who knows?

那么，从每个独立柱状结构随时间积累越来越多信息的角度来看，你觉得这种机制被充分理解了吗？

So in terms of each individual column building up more and more information over time, do you think that mechanism is well understood?

在你的想法中，你已经提出了很多种架构。

In in your mind, you've proposed a lot of architectures there.

这是关键部分吗？还是说，‘千脑智能理论’——所有这些的集合——才是核心？

Is that a key piece, or is it is the big piece the Thousand Brains Theory of Intelligence, the ensemble of it all?

嗯，我觉得两者都很重要。

Well, I think they're both big.

我的意思是，作为理论家，这个概念显然最令人兴奋。

I mean, clearly, concept as a theorist, the concept is most exciting.

对吧？

Right?

高级概念。

Have High level concept.

高级概念。

High level concept.

这是一种全新的思考方式，用来理解新皮层是如何工作的。

This is a totally new way of thinking about how the neocortex works.

所以这很有吸引力。

So that is appealing.

它带来了诸多影响。

It has all these ramifications.

以这个框架来理解大脑的工作方式，你可以做出各种预测并解决各种问题。

And with that as a framework for how the brain works, you can make all kinds of predictions and solve all kinds of problems.

现在我们正在努力梳理这些细节。

Now we're trying to work through many of these details right now.

好吧，神经元究竟是如何实现这一点的？

Okay, how do the neurons actually do this?

事实上，如果你思考大脑旧区域中的网格细胞和位置细胞，关于它们已经有很多研究，但仍有一些谜团。

Well, it turns out if you think about grid cells and place cells in the old parts of the brain, there's a lot that's known about them, but there's still some mysteries.

关于它们的具体工作机制和信号特征，目前还存在很多争议。

There's a lot of debate about exactly the details, how do these work and what are the signs?

我们同样面临着这种级别的细节和关注。

And we have that same level of detail, that same level of concern.

我们在这里花大部分时间做的事情，是尽可能详尽地列出我们还不理解的内容。

What we spend here most of our time doing is trying to make a very good list of the things we don't understand yet.

这才是关键所在。

That's the key part here.

有哪些限制条件？

What are the constraints?

并不是说，哦，这个东西似乎有效了，我们就完成了。

It's not like, Oh, this thing seems to work, we're done.

不，实际情况是，它勉强有效，但我们知道它还必须做到其他一些事情，而目前它还没做到。

No, it's like, Okay, it kind of works but these are other things we know it has to do and it's not doing those yet.

我认为我们在这方面已经取得了很大进展。

I would say we're well on the way here.

我们还没有完成。

We're not done yet.

这个系统有很多复杂之处，但关于新皮层不同层级如何完成大部分工作的基本原理，我们已经理解了。

There's a lot of trickiness to this system but the basic principles about how different layers in the neocortex are doing much of this, we understand.

但还有一些我们尚未充分理解的基本部分。

But there's some fundamental parts that we don't understand as well.

那么，你认为哪一个更难的开放性问题，或者哪一个一直让你困扰？

So what would you say is one of the harder open problems or one of the ones that have been bothering you?

哦，是的。

Oh, yeah.

最让你夜不能寐的是什么？

Keeping you up at night the most?

哦，好吧，现在这是一个细节问题，可能对大多数人不适用，好吧？

Oh, well, right now, this is a detailed thing that wouldn't apply to most people, okay?

当然。

Sure.

但你是想让我回答这个问题吗？

But you want me to answer that question?

是的，请说。

Yeah, please.

我们之前谈论过，好像要预测你对这个咖啡杯的感知，我需要知道我的手指在咖啡杯上的位置。

We've talked about as if, oh, to predict what you're going to sense on this coffee cup, I need to know where my finger's going to be on the coffee cup.

这没错，但还不够。

That is true, but it's insufficient.

想象一下我的手指触碰到咖啡杯的边缘。

Think about my finger touches the edge of the coffee cup.

我的手指可以以不同的角度触碰它。

My finger can touch it at different orientations.

我可以在这里旋转我的手指。

I can rotate my finger around here.

但这一点并不会改变。

And that doesn't change.

我依然能做出这样的预测。

I can make that prediction somehow.

所以这不仅仅是位置问题，还涉及方向成分。

So it's not just the location, there's an orientation component of this as well.

大脑的古老区域也有这种机制。

This is known in the old parts of the brain too.

有一种叫做头朝向细胞的东西，它们记录老鼠的朝向。

There's things called head direction cells, which way the rat is facing.

这是同一种基本理念。

It's the same kind of basic idea.

所以如果我的手指是一只老鼠，你知道，在三维空间中，我有三维的方向和三维的位置。

So if my finger were a rat, you know, in three dimensions, I have a three-dimensional orientation and I have a three-dimensional location.

如果我是一只老鼠，你可能会认为它有二维的位置或二维的方向，或者一维的方向，比如它只是朝哪个方向。

If I was a rat, I would have a, you might think of it as a two dimensional location or a two dimensional orientation, a one dimensional orientation, like just which way is it facing.

因此，这两个组成部分如何协同工作，我如何将传感器的方向与位置结合起来，这是一个棘手的问题。

So how the two components work together, how it is that I combine orientation, the orientation of my sensor as well as the location is a tricky problem.

我认为我在这一点上已经取得了进展。

And I think I've made progress on it.

所以从更大的角度来看，视角非常有趣但也非常具体。

So at a bigger version of that, so perspective is super interesting but super specific.

是的。

Yeah.

我提前警告过你。

I warned you.

不。

No.

这非常好，它有一个更通用的版本。

It's really good, it's a there's a more general version of that.

你觉得情境重要吗？比如我们身处北美的一栋建筑中，生活在有马克杯的时代？

Do you think context matters, the fact that we're in a building in North America that that we in the day and age where we have mugs?

我的意思是，你对房间里其他所有东西都带入了大量额外的信息，而不仅仅是咖啡杯。

I mean, there's all this extra information that you bring to the table about everything else in the room that's outside of just the coffee cup.

它是的。

Of it is.

它是如何获得的呢？

How does it get Yeah.

所以是关联的，

So connected,

你觉得呢？

do you think?