#99 – 卡尔·弗里斯顿：神经科学与自由能原理

所以我认为天气这个例子非常出色，你描绘的画面非常有趣。

So I think the the example of the weather is a brilliant one, and I think you paint a really interesting picture.

而且这并没有像我想象的那么消极。

And it wasn't as negative as I thought.

它本质上是在说，这可能极其具有挑战性，包括带宽的下限等等。

It's essentially saying that it's it might be incredibly challenging, including the low bound of the bandwidth and so on.

为了坦诚相告，我来自机器学习领域。

I kind of so just to full disclosure, I come from the machine learning world.

因此，我自然的想法是，最难的部分是工程上的挑战——如何将卫星送上天并让它们运行起来等等。

So my my natural thought is the hardest part is the engineering challenge of controlling the weather, of getting those satellites up and running and and so on.

一旦它们就位，剩下的就和让你在围棋中获胜的基本方法一样，也能让你在这片混沌的‘汤’中游刃有余。

And once they are, then the rest is fundamentally the same approaches that allow you to be to win in the game of Go will allow you to potentially play in this soup, in this chaos.

嗯。

Mhmm.

因此，我抱有希望，认为机器学习方法能帮助我们在这片混沌中发挥作用。

So I have I have a hope that sort of machine learning methods will will help us play in this soup.

但也许你说得对，这涉及生物学，大脑是一个极其复杂的系统，几乎难以介入。

But perhaps you're right that it is a biology and the brain is just an incredible an incredible system that may be almost impossible to get in.

但对我来说，真正看似不可能的是你所描述的那些错综复杂的血管系统，我们也不能忽视大脑的价值。

But for for me, what seems impossible is is the incredible mess of blood vessels that you also described without you know, we also value the brain.

你不能出任何差错。

You can't make any mistakes.

你不能损坏任何东西。

You can't damage things.

因此，对我而言，这个工程挑战几乎是不可能完成的。

So to me, that engineering challenge seems nearly impossible.

我在Neuralink最印象深刻的一点是，与那些杰出的神经外科医生和机器人专家交谈，让我意识到，尽管这看似不可能，但若有人能实现，那一定是这些正在努力攻克这一难题的世界级工程师。

One of the things I was really impressed by at Neuralink is just just just talking to brilliant neurosurgeons and the roboticists that made me realize that even though it seems impossible, if anyone can do it, it's some of these world class engineers that are trying to take it on.

所以，我认为我们这次讨论的结论是，这个问题确实非常困难，但或许并非完全不可能。

So so I think the conclusion of our discussion here is of of this part is is basically that the problem is really hard, but hopefully not impossible.

完全正确。

Absolutely.

那么，如果可以的话，我们从基础开始吧。

So if it's okay, let's start with the basics.

你也提出了一个非常有趣的原理——自由能原理。

So you've also formulated a fascinating principle, the free energy principle.

我们能不能先从基础开始，解释一下什么是自由能原理？

Can we maybe start at the basics and what is the free energy principle?

事实上，自由能原理继承了许多针对大脑产生的高维时间序列数据所发展的数据分析方法。

Well, fact, the free energy principle inherits a lot from the building of these data analytic approaches to these very high dimensional time series you get from the brain.

所以我认为承认这一点很有意义。

So I think it's interesting to acknowledge that.

特别是那些试图解决另一面——功能整合，也就是连接性分析——的工具。

And in particular, the analysis tools that try to address the other side, is the functional integration, so the connectivity analyses.

一方面，我也应该承认，它还大量借鉴了机器学习的成果。

On the one hand, but I should also acknowledge it inherits an awful lot from machine learning as well.

因此，自由能原理本质上是一个形式化的表述，它指出：任何能够在变化世界中存活的系统，其生存的基本要求都可以被看作是一个推断问题，即你可以将‘你存在的概率’理解为你存在的证据。

So the free energy principle is just a formal statement that the existential imperatives for any system that manages to survive in a changing world is can be cast as an inference problem in the sense that you can interpret the probability of existing as the evidence that you exist.

如果你能把存在的问题表述为一个统计问题，那么你就可以运用所有为推断而发展的数学工具，来理解和描述服务于这种推断的群体动力学。

And if you can write down that problem of existence as a statistical problem, then you can use all the maths that has been developed for inference to understand and characterize the ensemble dynamics that must be in play in the service of that inference.

因此，从技术上讲，这意味着任何因与环境分离而存在的事物，都可以被解释为试图最小化自由能的变异；如果你来自机器学习领域，你会知道这被称为负证据下界或负肘部，这等同于说你试图最大化或看起来所有动力学都在努力最大化其互补部分，即边际似然或你自身存在的证据。

So technically, what that means is you can always interpret anything that exists in virtue of being separate from the environment in which it exists as trying to minimize variation of free energy and if you're from the machine learning community, you will know that as a negative evidence lower bound or a negative elbow, which is the same as saying you're trying to maximize or it will look as if all your dynamics are trying to maximize the complement of that which is the marginal likelihood or the evidence for your own existence.

所以，这基本上就是所谓的自由能原理。

So that's basically the, you know, the free energy principle.

但为了稍微退后一步，你提到了‘存在性 imperative’。

But to even take a a sort of a small step backwards, you said the existential imperative.

这里有很多优美的诗意表达，但简单粗暴地说，这是一个非常有趣的想法，即当你观察一个团块时，你如何知道这个东西是活的？

There's a lot of beautiful poetic words here, but sort of to put it crudely, it's a it's a it's a fascinating idea of basically just of trying to describe if you're looking at a blob, how do you know this thing is alive?

嗯。

Mhmm.

什么是活着？

What does it mean to be alive?

存在意味着什么？

What does it mean to be to exist?

因此，你可以观察大脑，观察大脑的某些部分，或者这只是一个普遍原则，适用于几乎所有系统。

And so you can look at the brain, you can look at parts of the brain, or you this is just a general principle that applies to almost any system.

这仅仅是一个在哲学层面上令人着迷的问题，以及试图回答这个问题的方法——什么是活着？

That's just a fascinating sort of philosophically, at every level, question and the methodology to try to answer that question, what does it mean to be alive?

是的。

Yes.

所以这是一个庞大的工程，而从某种角度来看，至少有一个清晰的答案，这很好。

So so that that's a huge endeavor, and it's nice that there's at least some from some perspective, a clean answer.

那么，你能谈谈这种优化视角吗？

So maybe can you talk about that optimization view of it?

那么，究竟什么是被最小化或最大化的？

So what what's trying to be minimized to maximize?

一个有生命的系统，它试图最小化什么？

A system that's alive, what is it trying to minimize?

对。

Right.

你刚才做出了一个重大的转变。

You've you've made a big move there.

抱歉。

Apologize.

不。

No.

不。

No.

做出重大转变是好事。

It's it's good to make big moves.

但你假设了某物处于一种可能是生命或非生命的状态。

But you've assumed that the the thing exists in a state that could be living or non living.

所以我可能会问你，你凭什么说某物存在？

So I may ask you, what licenses you to say that something exists?

这就是我使用'存在主义'这个词的原因。

That's why I use the word existential.

它超越了生命。

It's beyond living.

它只是存在。

It's just existence.

所以，如果你深入探究存在的事物的定义，那么它们具有一些特定的属性——如果你借鉴非平衡稳态物理学中的数学方法，就可以将它们的存在解释为一种优化过程。

So if you drill down onto the definition of things that that exist, then they have certain properties if you borrow the maths from non equilibrium steady state physics that enable you to interpret their existence in terms of this optimization procedure.

很好，你提到了‘优化’这个词。

So it's good you introduce the word optimization.

因此，自由能原理在其最雄心勃勃但也最简约、最根本的表述中指出：如果某物存在，那么根据非平衡稳态的数学原理，它必然表现出某种特性，使其看起来像是在优化某个特定量；而这个特定量恰好与机器学习中的证据下界、贝叶斯统计中的贝叶斯模型证据完全一致——我还可以列出其他许多方式来理解这个关键量，它本质上是对意外、自信息的约束，如果你从信息论的角度来看的话。

So what the free energy principle in its sort of most ambitious but also most deflationary and simplest says is that if something exists, then it must by the mathematics of non equilibrium steady state exhibit properties that make it look as if it is optimizing a particular quantity and it turns out that particular quantity happens to be exactly the same as the evidence lower bound in machine learning or Bayesian model evidence in Bayesian statistics or and then I can list a whole other, you know, list of ways of understanding this this this key quantity which is a bound on on surprises, self information, if you information theory.

对于这个量，有多种不同的视角。

There are a whole there are a number of different perspectives on this quantity.

它本质上就是处于某种特定状态的对数概率。

It's just basically the log probability of being in a particular state.

我讲述这个故事，是真诚地试图回答你的问题，我以一个试图理解非平衡稳态基本原理的物理学家的身份来回答——但实际上我不该这么做，因为上一次我学习物理，还是在二十多岁的时候。

I'm telling this story as an honest attempt to answer your question and I'm answering it as if I was pretending to be a physicist who was trying to understand the fundamentals of non equilibrium steady state and I shouldn't really be doing that because the last time I was taught physics, was in my twenties.

当你想到自由能原理时，你想象的是哪种系统？作为一种更具体的案例研究，你指的是哪一类系统？

What kind of systems When you think about the free energy principle, what kind of systems are you imagining as a sort of more specific kind of case study?

是的。

Yeah.

我想象的是各种系统，但简单来说，就是一个能从其生态位或环境中识别出来的单细胞生物。

I'm imagining a range of systems, but yeah, at its simplest, a single celled organism that can be identified from its eco niche or its environment.

所以，最简单的情况就是我脑海中一直想象的那样。你可能会问，我们怎么可能对一滴油的存在提出任何问题呢？

So at its simplest, that's basically what I always imagined in my head And you may ask, well, is there any how on earth can you even elaborate questions about the existence of a single drop of oil for example?

是的。

Yeah.

但那里存在着深刻的问题。

But there are deep questions there.

为什么油不会溶解？为什么油滴内部与外部溶剂之间的界面能够持续存在？

Why doesn't the oil why doesn't the thing, the interface between the drop of oil that contains an interior and the thing that is not the drop of oil, which is the solvent in which it is immersed, how does that interface persist over time?

为什么油不会直接溶解到溶剂中？

Why doesn't the oil just dissolve into solvent?

那么，油滴表面与外部环境之间的交换具有什么特殊性质呢？如果物理学家说，这就是热路径。

So what special properties of the exchange between the surface of the oil drop and the external states in which it's immersed, if your physicists say, would be the heat path.

你知道，你面对的是一个物理系统，再次强调，我们讨论的是密度动力学、集合动力学，一群原子或分子浸没在热路径中。

You know, you've got a physical system, an ensemble again, we're talking about density dynamics, ensemble dynamics, an ensemble of atoms or molecules immersed in the heat path.

但问题是，热路径是如何出现的？为什么它没有被溶解？

But the question is, how did the heat path get there and why is it not dissolved?

它是如何维持自身的？

How is it maintaining itself?

没错。

Exactly.

它在进行什么行为？

What actions is it?

我的意思是，油滴这个想法太有趣了，我想它在水中会溶解。

I mean, it's such a fascinating idea of a drop of oil, I guess it would dissolve in water.

它不会在水中溶解。

It wouldn't dissolve in water.

具体是指什么？

What Precisely.

那为什么不呢？

So why not?

那为什么不呢？

So Why not?

为什么不呢？

Why not?

那你怎么用数学来描述呢？我的意思是，这个想法太美妙了，还有就是，这个油滴到底在哪里结束呢？

And and how do you mathematically describe I mean, it's such a beautiful idea, and also the idea of like where does the thing where does the drop of oil end Yeah.

而在哪里

And where does

对，没错。

it Right.

我的意思是，你问的是深刻的问题，不是那种肤浅的层面。

So I mean, you're asking deep questions deep in in a non millennial sense.

一种层级性的理解。

A hierarchical sense.

但你可以这样理解：这是它去膨胀化的一面。

But what you can do is So this is the deflationary part of it.

我能先澄清一下吗？通常当被问到这个问题时，我会从心理学家的角度来回答，也就是谈到预测处理、预测编码，以及大脑作为推理机器的概念。

Can I just qualify my answer by saying that normally when I'm asked this question, I answer from the point of view of a psychologist when we talk about predictive processing and predictive coding and you know, the brain as an inference machine?

但你并没有从这个角度问我，所以我现在是以物理学家的视角来回答。

But you haven't asked me from that perspective, I'm answering from the point of view of a physicist.

所以问题的重点不在于‘为什么’，而在于：如果它存在，它必须具备哪些特性？

So know, the question is not so much why, but if it exists, what properties must it display?

这就是自由能原理的去膨胀化部分。

So that's the deflationary part of the free energy principle.

自由能原理并不回答‘为什么’这个问题。

The free energy principle does not supply an answer as to why.

它说的是：如果某物存在，那么它就必须具备这些特性。

It's saying, if something exists, then it must display these properties.

这就是目前所提出的观点。

That's the sort of the thing that's on offer.

而恰好的是，这些它必须具备的特性非常引人入胜，并带有某种推断的色彩，一种自我验证的特质——这种特质源于这样一个事实：油滴与其非油滴之间边界的维持，需要优化某个特定函数或泛函，而这正是我最初从存在性命令谈起的原因。

And it so happens that these properties it must display are actually intriguing and have this inferential gloss, this sort of self evidencing gloss that inherits on the fact that the very preservation of the boundary between the oil drop and the not oil drop requires an optimization of a particular function or a functional that defines the presence of the existence of this oil drop, which is why I started with existential imperatives.

这种现象的发生是存在所必需的条件，因为边界本质上定义了存在的事物。

It is a necessary condition for existence that this must occur because the boundary basically defines the thing that's existing.

这就是自组织的方面。

So it is that self assembly aspect.

这正是你所暗示的：在生物学中，有时被称为自创生；在计算化学中，则称为自组装。

It's that you were hinting at In biology, sometimes known as autopoiesis, in computational chemistry, the self assembly.

它看起来像什么？

It's the does it look like?

抱歉。

Sorry.

你会如何描述那些从无到有自行构建的事物？

How would you describe things that configure themselves out of nothing?

它们清晰地与所处的环境或介质区分开来的方式。

The way they clearly demarcate themselves from the states or the soup in which they are immersed.

因此，从计算化学的角度来看，你可以将这理解为大分子的一种构型，以最小化其自由能，即热力学自由能。

So from the point of view of computational chemistry, for example, you just understand that as a configuration of a macro molecule to minimize its free energy, its thermodynamic free energy.

这正是我们之前讨论过的相同原理，热力学自由能其实就是负熵。

It's exactly the same principle that we've been talking about, that thermodynamic free energy is just the negative elbow.

这是一种相同的数学结构。

It's the same mathematical construct.

因此，存在、结构和形态的出现，即能够与环境或非自身事物区分开来的形式，必然要求存在一个目标函数，仿佛它在最小化这个函数。

So the very emergence of existence, of structure, of form that can be distinguished from the environment or the thing that is not the thing necessitates the existence of an objective function that it looks as if it is minimizing.

它在寻找自由能的最小值。

It's finding a free energy minima.

为了澄清一下，我正努力理解，自由能原理说的是，如果某物存在，它就应该具备这些特性。

And so just to clarify, I'm trying to wrap my my head around so the the free energy principle says that if something exists, these are the properties it should display.

是的。

Yeah.

所以这意味着我们不能只是简单地观察，不能直接进入这种汤状物中而没有任何机制。

So what it what that means is we can't just look, we can't just go into a soup and there's no mechanism.

自由能原理并没有给我们提供一种找到存在的事物的机制。

Free energy principle doesn't give us a mechanism to find the things that exist.

这是否意味着你可以用它来推理、思考，比如研究某个特定系统，并说：嘿，这个系统是否表现出这些特性？

Is is that what what's implying is being implied that you can kind of use it to reason, to think about, like, study a particular system and say, dude, does this exhibit these qualities?

这是个非常好的问题。

That's an excellent question.

但要回答这个问题，我必须回到你之前关于生命与非生命区别的问题。

But to answer that, I'd have to I have to return to your previous question about what's the difference between living and nonliving things.

嗯，实际上，不好意思。

Well, it's actually sorry.

所以，是的，也许我们可以往这个方向探讨。

So, yeah, maybe we can go there.

你画了一条线，原谅我问些愚蠢的问题，但你似乎在生命和存在之间画了一条界线。

You kind of drew a line, and and forgive me for the stupid questions, but you kinda drew a line between living and existing.

是的

Yeah.

是否存在一个有趣的区别？

Is there an interesting sort of Distinction.

两者之间的区别。

Distinction between the two.

我认为是有的。

I I think there is.

你知道，事物确实存在，比如沙粒、月球上的岩石、树木，还有你。

So, you know, things do exist, grains of sand, rocks on the moon, trees, you.

所有这些事物都可以与它们所处的环境区分开来，因此，它们在某种程度上一定在优化自身的自由能，从这种模型证据的视角来看，这意味着它们是自我确证的。

So all of these things can be separated from the environment in which they are immersed and therefore, they must at some level be optimizing their free energy, taking this sort of model evidence interpretation of this quantity that basically means they're self evidencing.

这里还有一个巧妙的说法：从统计学角度来说，你就是你自身存在的证明，我不记得自己说过这句话，但有人说过，我非常喜欢这个说法。

Another nice little twist of phrase here is that you are your own existence proof, you know, statistically speaking, which I don't think I said that, somebody did, but I love that phrase.

你就是你自身存在的证明。

You are your own existence proof.

是的。

Yeah.

所以这太存在主义了，不是吗？

So it's so existential, isn't it?

它是

It's

我得花几天时间好好想想。

I'm gonna have to think about that for a few days.

对。

Yeah.

这是一句美妙的话。

That's beautiful line.

所以，要回答你关于它有什么用的问题，可以沿着以下思路来思考。

So the step through to answer your question about what's it good for, go along the following lines.

首先，你必须定义什么是存在，正如你正确指出的，你必须界定某物的状态必须具备哪些概率特性，才能知道它的边界在哪里。

First of all, you have to define what it means to exist, which now as you've rightly pointed out, you have to define what probabilistic properties must the states of something possess so that it has so it knows where it finishes.

然后你将这些用统计独立性来表达，同样涉及稀疏性。

And then you write that down in terms of statistical independences, again, sparsity.

再次强调，关键不在于什么相互连接、什么相关或什么依赖于什么。

Again, it's not what's connected or what's correlated or what depends upon what it is.

而在于什么不相关、什么不依赖于其他事物。

It's what's not correlated and what doesn't depend upon something.

同样，这归结为深层结构，这次不是层级结构，而是通过消除连接和依赖所涌现出来的结构。

Again, it comes down to the deep structures, not in this instance hierarchical, but the structures that emerge from removing connectivity and dependency.

在这种情况下，基本上就是能够从其所浸没的水中识别出油滴的表面。

In this instance, basically being able to identify the surface of the oil drop from the water in which it is immersed.

当你这样做时，你会开始意识到，在任何包含事物的宇宙中，实际上存在四种状态。

And when you do that, you start to realize, well, there are actually four kinds of states in any given universe that contains anything.

一种是表面内部的事物，一种是表面外部的事物，还有表面本身——这正是我使用隐喻的原因：一个单细胞生物，拥有内部、外部以及细胞表面。

The things that are internal to the surface, the things that are external to the surface and the surface in and of itself, which is why I use a metaphor, a little single celled organism that has an interior and exterior and then the surface of the cell.

而这在数学上就是马尔可夫毯。

And that's mathematically a Markov blanket.

让我停一下，我对这个概念感到无比震撼：有表面之外的东西、表面之内的东西，以及表面本身——也就是马尔可夫毯。

Just to pause, I'm in awe of this concept that there's the stuff outside the surface, stuff inside the surface, and the surface itself, the Markov blanket.

这真是一个极其美妙的概念，试图探索存在本身意味着什么。

It's just the most beautiful kind of notion about trying to explore what it means to exist automatically.

对不起。

I I apologize.

这真是一个非常美妙的想法。

It's just a it's just a beautiful idea.

但这个概念源自加州，所以那就是

But it came out of California, so that's

我改变主意了。

I changed my mind.

我收回刚才所有的话。

I take it all back.

所以，总之，你刚才谈到的表面，关于马尔可夫毯。

So so anyway, so what you were just talking about the surface, about the Markov Yeah.

所以这个表面或这个毯子，这些毯子状态，之所以如此，是因为它们现在是根据这些独立性来定义的——即内部状态、毯子状态或外部状态中，哪些可以相互影响，哪些不能相互影响。

So this surface or this blanket these blanket states that are this you know, the the because they are now defined in relation to these independencies and what different states internal or blanket or external states can which ones can influence each other and which cannot influence each other.

现在你可以应用非平衡物理、稳态、热力学或流体动力学中常见的标准结果，将它们应用于这种划分。

You can now apply standard results that you would find in non equilibrium physics or steady state or thermodynamics or hydrodynamics, usually out of equilibrium solutions and apply them to this partition.

看起来，如果你将任何非平衡系统所关联的正常梯度流以某种方式应用，那么马克ov毯的两个部分和内部状态似乎都在对同一个量进行爬山或梯度下降。

What it looks like is if all the normal gradient flows that you would associate with any non equilibrium system apply in such a way that two part of the Markov blanket and the internal states seem to be hill climbing or doing a gradient descent on the same quantity.

这意味着你现在可以用流、动力学、运动方程来描述这个油滴的存在。

And that means that you can now describe the very existence of this oil drop.

你可以用流、动力学、运动方程来描述这个油滴的存在，其中毯子状态的一部分——我们称之为活性状态——和内部状态现在似乎必须努力看起来像是在最小化同一个函数，即最大化占据这些状态的概率。

You can write down the existence of this oil drop in terms of flows, dynamics, equations of motion where the blanket states or part of them, we call them active states and the internal states now seem to be and must be trying to look as if they're minimizing the same function, which is a lot of probability of occupying these states.

有趣的是，如果你试图描述这些事物，它们该被称作什么？

The interesting thing is that what would they be called if you were trying to describe these things?

所以我们讨论的是内部状态、外部状态和毯子状态。

So what we're talking about are internal states, external states and blanket states.

现在让我们把毯子状态划分为两类：感觉状态和活性状态。

Now let's carve the blanket states into two, sensory states and active states.

从操作上讲，为了使这种对不同状态集合的划分成为可能，主动状态和马尔可夫毯子不能受到外部状态的影响。

Operationally, it has to be the case that in order for this carving up into different sets of states to exist, the active states, the Markov blanket cannot be influenced by the external states.

我们已经知道，内部状态不会受到外部状态的影响，因为马尔可夫毯子将它们隔开了。

And we already know that the internal states can't be influenced by the external states because the blanket separates them.

那这意味着什么？

So what does that mean?

这意味着，主动状态和内部状态共同不受外部状态的影响。

Well, it means the active states, the internal states are now jointly not influenced by external states.

它们只具有自主的动力学。

They only have autonomous dynamics.

于是，你现在有了一个具有自主性的油滴的图景。

So now you've got a picture of an oil drop that has autonomy.

它拥有自主的状态。

It has autonomous states.

它拥有自主的状态，意思是油滴表面的某些部分以及内部都不受外部状态的影响。

It has autonomous states in the sense that there must be some parts of the surface of the oil drop that are not influenced by the external states and all the interior.

而这两种状态共同赋予了哪怕是一个小小的油滴以看似在优化其自由能或负熵、其模型证据的自主状态。

And together, those two states endow even a little oil drop with autonomous states that look as if they are optimizing their variation free energy or their negative elbow, their model evidence.

这将是一个有趣的智力练习，你甚至可以进入泛心论的领域，认为所有存在的事物都在隐式地进行自我证成的推断。

That would be an interesting intellectual exercise and you could say, you could even go into the realms of panpsychism that everything that exists is implicitly making inferences on self evidencing.

现在，我们进一步推进，但生命体又如何呢？

Now, we make the next move, but what about living things?

我问你，油滴和一只小蝌蚪、小幼虫或浮游生物之间有什么区别？

Let me ask you, what's the difference between an oil drop and a little tadpole or a little larva or a plankton?

我们刚刚描绘的油滴图景，仅仅几分钟内就让我进入了泛心论的世界，你让我相信油滴确实是一种有生命、至少是自主的系统。

The picture we just painted of an oil drop just immediately in a matter of minutes took me into the world of panpsychism, where you you just convinced me made me feel like an oil drop is a living, certainly an autonomous system, but almost a living system.

它具备感知能力和行动能力，并能维持某种状态。

So it has it has a capability sensory capabilities and acting capabilities and and maintain something.

那么，这和我们传统上认为的生命系统之间有什么区别呢？

So what is the difference between that and something that we traditionally think of as a living system?

那就是它会死或者不会死，我的意思是，死亡。

That it could die or it can't, I mean, yeah, mortality.

我不太确定。

I'm not I'm not exactly sure.

我不确定正确的答案是什么，因为它们能够移动，而运动似乎是能够对环境施加作用的关键要素，但油滴也在做同样的事。

I'm not sure what the right answer there is, because they can move like, movement seems like an essential element to being able to act in the environment, but the oil drop is doing that.

所以我不知道。

So I don't know.

是吗？

Is it?

油滴是被推动的，但它自身是否能自主移动？

The oil drop will be moved, but does it in and of itself move autonomously?

嗯，表面在执行一些维持其结构的动作。

Well, well, the surface is performing actions that maintain its its structure.

是的。

Yeah.

你太机灵了。

You're being too clever.

我那时很难过。

I was I was sad.

我不介意一个安静地躺在那里的小油滴。

I didn't mind a passive little oil drop that's sitting there

是的。

Yeah.

在一杯水的底部。

At the bottom of the top of a glass of water.

当然。

Sure.

我想是的。

I guess

我想表达的是，你完全正确。

What I'm trying to say is you're absolutely right.

你说到点子上了。

You've nailed it.

这是一种运动。

It's movement.

是的。

Yeah.

那么这种运动从何而来？

So where does that movement come from?

如果它来自内部，那么我认为这就是某种有生命的东西。

If it comes from the inside, then then you've got I think something that's living.

你所说的'来自内部'是什么意思？

What do you mean from the inside?

我的意思是，内部状态可以影响主动状态，而主动状态可以影响外部状态，但外部状态却无法影响它们，正是这些内部状态引发了运动。

What I mean is that the internal states that can influence the active states where the active states can influence but they're not influenced by the external states can cause movement.

所以，如果你愿意这么说的话，油滴有两种类型。

So there are two types of oil drops if you like.

有一种油滴，其内部状态过于随机，以至于相互抵消了。

There are oil drops where the internal states are so random that they average themselves away

我明白了。

I see.

从整体和平均来看，当进行平均化时，这种东西是无法移动的。

And the thing cannot on balance on average when you do the averaging move.

一个很好的例子就是太阳。

So a nice example of that would be the Sun.

太阳当然具有内部状态，存在大量内在的自主活动，但由于这些活动没有协调性，缺乏像大脑那样的深层、千年尺度的层级结构，因此没有任何整体模式、规律或组织能在表面表现出来，使其真正地游动。

The sun certainly has internal states, there's lots of intrinsic autonomous activity going on, but because it's not coordinated, because it doesn't have the deep in the millennial sense hierarchical structure that the brain does, there is no overall mode or pattern or organization that expresses itself on the surface that allows it to actually swim.

它的表面当然可以非常活跃，但从整体尺度上看，太阳表面的平均位置本身无法移动，因为其内部动力学更像是一种高温气体。

It can certainly have a very active surface, but on mass, at the scale of the actual surface of the Sun, the average position of that surface cannot in itself move because the internal dynamics are more like a hot gas.

它们本质上就是高温气体。

They are literally like a hot gas.

而你的内部动力学则更加结构化、深度结构化，你现在可以通过肌肉、分泌器官、自主神经系统及其效应器来表达你的标记和主动状态。

Whereas your internal dynamics are much more structured and deeply structured and now you can express on your mark off and your active states with your muscles and your secretory organs, your autonomic nervous system and its effectors.

你确实能够移动，而这正是你唯一能做的事。

You can actually move and that's all you can do.

这是一件你如果以前没有这样想过的话，会觉得很有趣的事情：改变宇宙的唯一方式，就是移动。

That's something which you know, if you haven't thought of it like this before, I think it's nice to just realize there is no other way that you can change the universe other than simply moving.

无论是通过声带发出声音、四处走动，还是通过分泌器官排出分泌物，你改变宇宙的唯一方式，就是移动。

Whether that moving is articulating my with my voice box or walking around or squeezing juices out of my secretory organs, there's only one way you can change the universe, It's moving.

而你并非随机地这样做，这正是你拥有生命的标志。

And and the fact that you do so non randomly makes you alive.

是的。

Yeah.

正是这种非随机性。

So it's that non randomness.

所以这就是关键所在，它体现为本质上是游泳、本质上是移动、改变形态，一种动态且可能具有适应性的形态发生。

So the the that's what so it is and that would be manifested, we realized in terms of essentially swimming, essentially moving, changing one shape, a morphogenesis that is dynamic and possibly adaptive.

这就是我试图阐明油滴和小蝌蚪之间的区别。

So that's what I was trying to get out between the difference from the oil drop and the little tadpole.

小蝌蚪是在游动的。

The tadpole is moving around.

活跃状态实际上在改变外部状态，并且现在存在一个行动-感知循环，或者说一种反复的动态过程，这种过程依赖于这种深深结构化的自主行为，而这种行为建立在内部动力学之上，这些动力学不仅建模了作用于其表面的数据或整体状态，还通过移动主动重新采样这些数据。

Active states are actually changing the external states and there's now a cycle, an action perception cycle if you like, a recurrent dynamic that's going on that depends upon this deeply structured autonomous behavior that rests upon internal dynamics that are not only modeling the data impressed upon their surface or the blanket states, but they are actively resampling those data by moving.

它们正在向化学梯度和趋化性移动。

They're moving towards chem say chemical gradients and chemotaxis.

因此，它们已经超越了仅仅成为很好地模拟其所处世界类型的模型。

So they've gone beyond just being good little models of the kind of world they live in.

例如，从泛灵论的角度来看，一个油滴可以被看作是一个小生命体，它已经完美地推断出自己是一个被动的、非生命的油滴，生活在一盆水里。

For example, an oil droplet could in a panpsychic sense be construed as a little being that has now perfectly inferred its passive non living oil drop living in a bowl of water.

这没有问题。

No problem.

但若要赋予这个油滴能力，使其能够出去检验关于不同存在状态的假设，比如实际推动其表面到那里、再那里，以探测化学梯度，那么你就开始迈向更接近生命的形式。

But to now equip that oil drop with the ability to go out and test that hypothesis about different states of beings, so it can actually push its surface over there, over there and test for chemical gradients or then you start to move to much more lifelike form.

这在理论上很有趣，但事实上，它对于反映我在千禧年之后所观察到的现象至关重要，那就是向一种非被动的、具身化的智能理解转变。

This is fun, theoretically interesting, but it actually is quite important in terms of reflecting what I have seen since the turn of the millennium, which is this move towards an inactive and embodied understanding of intelligence.

你说你来自机器学习。

You say you're from machine learning.

是的。

Yes.

这意味着，运动的核心重要性，我认为尚未真正影响到机器学习。

So what that means, this this sort of the central importance of movement, I think is yet to really hit machine learning.

它当然已经渗透到机器人领域，也许在某些主动视觉问题中也是如此，因为你必须移动摄像头来采样这些和那些信息。

It certainly has now diffused itself throughout robotics and perhaps you could say certain problems in active vision where you actually have to move the camera to sample this and that.

但数据挖掘和深度学习类型的机器学习尚未应对这一问题。

But machine learning of the data mining deep learning sort simply hasn't contended with this issue.

它没有去处理运动问题和数据的主动采样，而是直接说：我们不需要担心它。

What it's done instead of dealing with the movement problem and the active sampling of data, it's just said, we don't need to worry about it.

我们可以看到所有数据，因为我们有大数据。

We can see all the data because we've got big data.

所以我们不需要关注运动。

So we need ignore movement.

因此，对我来说，这是当前机器学习中的一个重要疏漏。

So that for me is an important omission in current machine learning.

当前的机器学习更像油滴。

The current machine learning is much more like the oil drop.

是的。

Yes.

但这是一个享受接触几乎全部数据的油滴，是的。

But an oil drop that enjoys exposure to nearly all the data Yes.

需要被接触。

Need to be exposed to.

而不是蝌蚪游出去寻找合适的数据。

As opposed to the tadpoles swimming out to find the right data.

例如，它喜欢食物。

For example, it likes food.

这是个不错的假设。

That's a good hypothesis.

我们来验证一下。

Let's test it out.

让我们去行动、移动并摄取食物，比如，看看这是否能证明我这种东西喜欢这类食物。

Let's go and move and ingest food, for example, and see what that you know, is that evidence that I'm the kind of thing that likes this kind of food.

所以下一个自然的问题是，请原谅我提出这个问题，但如果我们设想一下人工智能系统，我刚刚描绘了一幅关于存在和生命的美好图景。

So the the next natural question, and forgive this question, but if we think of sort of even artificial intelligence systems, which I just painted a beautiful picture of existence and life.

那么，在这个框架中，你是否认为有可能定义意识，或者探索意识这个概念？

So do you you ascribe do do you find within this framework a possibility of defining consciousness or exploring the idea of consciousness?

比如，自我意识，嗯。

Like, what, you know, self awareness Uh-huh.

并将其扩展到意识层面，是的。

And expand it to consciousness, like, yeah.

我们该如何在这个框架中开始思考意识？

How can we how can we start to think about consciousness within this framework?

这有可能吗？

Is it possible?

嗯，这是有可能的。

Well, it's yeah.

我认为我们可以思考这个问题，但你是否能取得进展才是关键。

I think it's possible to think about it whether you'll get

能否有所突破才是问题所在。

Get anywhere is the question.

而且，我不确定自己是否有资格回答这个问题。

And again, I'm not sure that I'm licensed to answer that question.

我认为你得去问问专业的哲学家，才能得到一个明确的答案。

I I think you'd have to speak to a qualified philosopher to get a definitive answer there.

但毫无疑问，目前人们对利用这些理念，以及信息论中相关概念，来试图界定意识的数学、微积分和几何结构有着浓厚兴趣，无论是从最小意识，甚至比最小自我意识更初级的层面入手。

But certainly, there's a lot of interest in using not just these ideas but related ideas from information theory to try and tie down the maths and the calculus and the geometry of consciousness either in terms of sort of a minimal consciousness, even less than a minimal selfhood.

我所指的是有效规划和具备能动性的能力。

What I'm talking about is the ability effectively to plan, to have agency.

因此，你可以说病毒具有一种能动性，因为它能选择性地寻找宿主和细胞来寄生并四处移动。

So you could argue that a virus does have a form of agency in virtue of the way that it selectively finds hosts and cells to live in and moves around.

但你不会认为它具备思考规划和有目的性地移动的能力，更不会认为它能预见未来。

But you wouldn't endow it with the capacity to think about planning and moving in a purposeful way where it countenances the future.

但你可能会认为蚂蚁不是像病毒那样完全无意识的。

Whereas you might an ant.

你可能会觉得蚂蚁的无意识程度不如病毒那么高。

You might think an ant's not quite as unconscious as a virus.

它显然似乎有某种目的。

It certainly seems to have a purpose.

它在觅食途中会与同伴交流。

It talks to its friends en route during its foraging.

它拥有另一种形式的自主性，这种自主性是生物性的，且超越了病毒。

It has a different kind of autonomy, which is biotic but beyond a virus.

所以，关于规划的复杂性，确实存在一条界限，对吧？

So there's something about so there's some line that has to do with the complexity of planning Yes.

这可能包含了一个答案。

That may contain an answer.

我的意思是，如果我们能找出一条界限，超过这条界限就可以说一个生物是有意识的，那该有多美好。

I mean, it'd be beautiful if if we can find a line beyond which we could say a being is conscious.

是的。

Yes.

将会是的。

It will be.

对。

Yeah.

这些是我们用存在、生命和意识所划定的美妙界限。

These are wonderful lines that we've drawn with existence, life, and consciousness.

是的。

Yes.

那将会非常美好。

It will be very nice.

那里有一个小小的细节，这是我过去几个月才了解到的，那就是模糊性的概念。

One little wrinkle there, and this is something I've only learned in the past few months, is the notion of vagueness.

所以你是说，划定一条界限将会非常美好。

So you're saying it would be wonderful to draw a line.

我一直以为那条线终归会被画出来，直到大约四个月前，一位哲学家向我介绍了模糊性的概念。

Had always assumed that that line at some point would be drawn until about four months ago, and the philosopher taught me about vagueness.

所以我不知道你是否接触过这个概念，但这是一个技术性概念，我认为最生动的例证是：一堆沙子在什么时候才算是一堆？

So I don't know if you've come across this, but it's a technical concept and I think most revealingly illustrated with at what point does a pile of sand become a pile?

是一粒沙子、两粒沙子、三粒沙子，还是四粒沙子？

Is it one grain, two grains, three grains or four grains?

那么，你在什么时候会划出一条线，来区分‘一堆沙子’和‘沙粒的集合’？

So at what point would you draw the line between being a pile of sand and a collection of grains of sand?

同样地，我们是否应该追问：在意识与无意识之间，我该在哪里划出界限？

In the same way, is it right to ask where would I draw the line between conscious and unconscious?

这可能是一个模糊的概念。

And it might be a vague concept.

话虽如此，我完全同意你的观点。

Having said that, I agree with you entirely.

那些具备规划能力的系统，从技术上讲，这意味着你的推理自我证据过程——我简单地指代支撑你保持油滴状形态的热力学与梯度流，这些可以被描述为对对数贝叶斯模型证据的优化，你的肘部。

Systems that have the ability to plan, So just technically what that means is your inferential self evidencing by which I simply mean the dynamics, literally the thermodynamics and gradient flows that underwrite the preservation of your oil droplet like form are described as a can be described as an optimization of log Bayesian model evidence, your elbow.

这种自我验证必然是对导致你感官表面或标记毯上感官印象之原因的模型的证据。

That self evidencing must be evidence for a model of what's causing the sensory impressions on the sensory part of your surface or your mark off blanket.

如果这个模型具备规划能力，它就必须包含对你主动状态或行为的未来后果的模型，也就是规划本身。

If that model is capable of planning, it must include a model of the future consequences of your active states or your action, just planning.

所以我们现在进入了将规划视为推断的领域。

So we're now in the game of planning as inference.

但现在请注意我们已经做出了怎样的转变。

Now notice what we've made though.

我们已经大大偏离了大数据和机器学习，因为这再次是迁移所带来的后果。

We've made quite a big move away from big data and machine learning because again, it's the consequences of moving.

这是选择那些数据、那些数据，或往那边看的后果。

It's the consequence of selecting those data or those data or looking over there.

这立即表明，即使要成为有意识的人工智能、强人工智能或通用人工智能的候选者，

That tells you immediately that even to be a contender for a conscious artifact or strong AI or generalized

通用人工智能。

General AI.

那么，你必须在这一过程中引入行动。

Then you've got to have movement in the game.

此外，你还需要一个生成模型，比如变分自编码器那样的模型，它能根据不同的行动路径来预测未来。

Furthermore, you've got to have a generative model of the sort you might find in say a variation autoencoder that is thinking about the future conditioned upon different courses of action.

这带来了一系列关键要素，让你开始觉得，这些要素确实具备了讨论意识的所有必要条件。

That brings a number of things to the table which now you start to think, Well, those have got all the right ingredients to talk about consciousness.

现在，我必须在多种可能的未来行动路径中做出选择，这是规划的一部分。

I've now got to select among a number of different courses of action into the future as part of planning.

我现在拥有了自由意志。

I've now got free will.

选择这一行动路径、这一策略或那一行动，突然间让我成为了一个推理机器，一个自我验证的实体，它似乎在主动地游动、游向这里或那里，或看向这里或那里时，在多种可能的前进方式之间做出选择。

Act of selecting this course of action or that policy or that policy or that action suddenly makes me into an inference machine, a self evidencing artifact that now looks as if it's selecting amongst different alternative ways forward as I actively swim here or swim there or look over here or look over there.

我认为，一旦规划被纳入其中，你就已经非常接近那条界限了——如果那条界限确实存在的话。

I think you've now got to a situation if there is planning in the mix, you're now getting much closer to that line if that line were ever to exist.

但我认为这还未能让你真正达到自我意识的层面。

I don't think it gets you quite as far as self aware though.

我认为你必须面对这个问题：如何形式化地写出自我意识的微积分或数学？

I think you And then you're You have to I think grapple with the question, how would formally write down calculus or a maths of self awareness?

我认为这并非不可能，但你可能会面临压力，必须对‘自我意识’的含义给出一个正式的定义。

I don't think it's impossible to do, but I think you would There will be pressure on you to actually commit to a formal definition of what you mean by self awareness.

我认为我认识的大多数人可能会说，一条金鱼、一只宠物鱼没有自我意识，他们可能会对自家最爱的猫是否具有自我意识有争议，但会很乐意说他们的母亲是有自我意识的。

I think most people that I know would probably say that a goldfish, a pet fish was not self aware, they would probably argue about their favorite cat, but would be quite happy to say that their mom was self aware.

嗯。

Mhmm.

我的意思是，这很可能与某种层次的规划复杂性有关。

I mean, but that might very well connect to some level of complexity with planning.

看起来自我意识是复杂规划所必需的。

It seems like self awareness is essential for complex planning.

是的。

Yeah.

你想进一步探讨这个观点吗？

Do you want to take that further?

因为我觉得你说得完全正确。

Because I think you're absolutely right.

再次强调，这条界限并不清晰，但似乎将你自己融入世界、融入你的规划中，对于构建复杂计划是至关重要的。

Again, the line is unclear, but it seems like integrating yourself into the world, into your into your planning is essential for constructing complex plans.

是的。

Yes.

对。

Yeah.

用与自由能原理相同优雅的数学方式来描述这一点，可能会很困难。

By mathematically describing that in the same elegant way as you have with the free energy principle might be difficult.

嗯，也不完全是。

Well, yes and no.

我不认为我们该继续下去，不如我们先回退一步？

I don't think that well, perhaps we should just can we just go back?

你刚才给出的回答非常重要，我想如果我把它拆解开来，你会看到你刚刚向我们揭示的那些不言而喻的真理。

That's a very important answer you gave, and I think if I just unpacked it, you'd see the truisms that you've just exposed for us.

抱歉，我意识到之前没有回答你的问题。

Let me Sorry, I'm mindful that I didn't answer your question before.

自由能原理有什么用？

What's the free energy principle good for?

它只是用来解释非平衡稳态的一种漂亮的理论练习吗？

Is it just a pretty theoretical exercise to explain non equilibrium steady state?

是的，确实如此。

Yes it is.

它除了这一点之外对你没有任何帮助。

It does nothing more for you than that.

它可以被看作——听起来可能很自大，但它类似于自然选择的理论或自然选择的假设。

It can be regarded, it's going sound very arrogant, but it is of the sort of theory of natural selection or hypothesis natural selection.

很优美，毫无疑问是正确的，但它完全无法解释你为什么有腿和眼睛。

Beautiful, undeniably true, but tells you absolutely nothing about why you have legs and eyes.

它对实际表型没有任何说明，也无法让你据此构建任何东西。

It tells you nothing about the actual phenotype and it wouldn't allow you to build something.

自由能本身就像大多数同义反复的理论一样空洞。

The free energy by itself is as vacuous as most tautological theories.

所谓同义反复，我指的是自然选择理论，即适者生存。

By tautological, of course, I'm talking to the theory of natural the survival of the fittest.

适者中的‘适’指的是什么？

What's the fittest of the survival?

为什么它们会趋向更适应者？

Why do they cycle to the fitter?

这只不过是循环论证。

It just goes around in circles.

从某种意义上说，自由能原理背后也有着同样的去实质化的同义反复。

In a sense, the free energy principle has that same deflationary tautology under the hood.

这是存在之物的一个特征。

It's a characteristic of things that exist.

它们为何存在？

Why they exist?

因为它们最小化了自由能。

Because they minimize their free energy.

为什么它们要最小化自由能？

Why they minimize their free energy?

因为它们存在。

Because they exist.

你只是不停地绕圈、绕圈、再绕圈。

You just keep on going round and round and round.

自然选择无法提供这种实际效果，但你现在可以在诸如差分进化、遗传算法或机器学习中的MCMC等例子中看到它显现出来。

The practical thing which you don't get from natural selection, but you could say has now manifest in things like differential evolution or genetic algorithms or MCMC for example in machine learning.

你能获得的实际效果是：如果看起来存在之物似乎在努力实现一种密度动态，这种动态仿佛在优化变分自由能，而变分自由能必须是生成模型的函数——即对原因与结果的概率性描述，其中原因存在于外部，结果体现在马克ov毯的感觉部分——那么理论上，你应该能够写出这个生成模型，计算出梯度，从而使其自主地自我证实。

The practical thing you can get is if it looks as if things that exist are trying to have density dynamics that look as if they're optimizing a variational free energy, and a variation of free energy has to be a function of a generative model, a probabilistic description of causes and consequences, causes out there, consequences in the sensorium on the sensory parts of the Markov blanket, then it should in theory be possible to write down the generative model, work out the gradients and then cause it to autonomously self evidence.

所以，你应该能够写出油滴的模型。

So you should be able to write down oil droplets.

你应该能够创造出人工制品，在其中你提供了目标函数，该函数提供梯度和自组织动力学，使其达到非平衡稳态。

You should be able to create artifacts where you have supplied the objective function that supplies the gradients and supplies the self organizing dynamics to non equilibrium steady state.

因此，当你能够以某种方式写出所需的证据，即当你能够写出那个包含证据的生成模型时，自由能原理实际上就有了实际应用。

So there is actually a practical application of free energy principle when you can write down your required evidence in terms of well, when you can write down the generative model that is the thing that has the evidence.

给定该模型的这些感官数据或这些数据的概率，实际上就是肘部或变分自由能所界定或近似的东西。

The probability of these sensory data or this data given that model is effectively the thing that the elbow or the variation free energy bounds or approximates.

这意味着你可以实际写出这个模型，以及你想要工程化的那种你希望以概率方式实现的AGI或通用人工智能，然后你进行大量的艰苦工作，但你会设计一个机器人和计算机来对该目标函数执行梯度下降。

That means that you can actually write down the model and the kind of thing that you want to engineer, the kind of AGI or artificial general intelligence that you want to manifest probabilistically, and then you engineer, a lot of hard work, but you would engineer a robot and a computer to perform a gradient descent on that objective function.

因此，它确实具有实际意义。

So it does have a practical implication.

那么，我为什么一直在谈论这个呢？

Now why am I wittering on about that?

这看起来确实与是的有关。

It did seem relevant to yes.

那么，答案是容易还是困难呢？

So what kinds of so the answer to would it be easy or would it be hard?

嗯，从数学上讲，这很容易。

Well, mathematically, it's easy.

我刚刚告诉过你们，你们要做的就是以概率生成模型的形式，概率性地写出你们完美的产物，即对这个事物所处世界中的因果关系的概率分布，然后只需设计一台计算机和一个机器人，对这个目标函数进行梯度下降。

I've just told you all you need to do is write down your perfect artifact, probabilistically in the form of a probabilistic generative model, probability distribution over the causes and consequences of the world in which this thing is immersed, and then you just engineer a computer and a robot to form a gradient descent on that objective function.

没问题。

No problem.

但当然，主要的问题在于写出这个生成模型。

But of course, the big problem is writing down the generative model.

所以，真正的难点就在这里。

So that's where the heavy lifting comes in.

因此，正是这个生成模型的形式与结构，从根本上定义了你们将要创造的产物，或者说具有自我意识的产物类型。

So it's the form and the structure of that generative model which basically defines the artifact that you will create or indeed the kind of artifact that has self awareness.

所以，所有的艰苦工作都集中在这里。

So that's where all the hard work comes.

就像自然选择丝毫没有告诉你为什么你会有眼睛一样，你必须深入研究实际的表型、实际的生成模型。

Very much like natural selection doesn't tell you in the slightest why you have eyes, so you have to drill down on the actual phenotype, the actual genitive model.

鉴于此，你之前告诉我什么，能让我立刻明白为了实现自我意识，我必须写出什么样的生成模型？

With that in mind, what did you tell me that tells me immediately the kinds of genitive models I would have to write down in order to have self awareness?

你告诉我的是，我必须拥有一个适用于我所处的这种世界的模型。

What you said to me was I have to have a model that is effectively fit for purpose for this kind of world in which I operate.

如果我现在观察到，这种世界本质上是由许多像我这样的事物构成的，即‘我’。

And if I now make the observation that this kind of world is effectively largely populated by other things like me, I.

你。

E.

那么，如果我能提出一个假设，认为我们是相似的生物，事实上是同一种生物，只是我是我，你是你，那么就必然需要一种自我意识。

You, then it makes enormous sense that if I can develop a hypothesis that we are similar kinds of creatures, in fact, the same kind of creature, but I am me and you are you, then it becomes again mandated to have a sense of self.

因此，如果我生活在一个由像我这样的事物构成的世界中，本质上是一个社会性世界、一个社群，那么我就必须推断出，说话的是我，而不是你。

So if I live in a world that is constituted by things like me, basically a social world, a community, then it becomes necessary now for me to infer that it's me talking and not you talking.

如果我独自一人在火星上，或者作为一个野孩子生活在丛林里，我就不会需要这种推断。

I wouldn't need that if I was on Mars by myself or if I was in the jungle as a feral child.

如果周围没有任何像我这样的存在，那就没有必要推断出：啊，是我在体验或引发这些声音，而不是你。

There was nothing like me around, there would be no need to have an inference that a hypothesis, Ah yes, it is me that is experiencing or causing these sounds and it is not you.

只有当这个世界中存在其他个体，从而带来模糊性时，这才变得必要。

It's only when there's ambiguity in play induced by the fact that there are others in that world.

所以我认为，自我意识人工制品的特别之处在于，它们已经学会、获得，或者至少可能通过进化具备了生成模型，能够应对周围存在大量与它们相似的实体这一事实，因此它们必须分辨出‘那是你，不是我’。

So I think that the special thing about self aware artifacts is that they have learned to or they have acquired or at least are equipped with possibly by evolution generative models that allow for the fact there are lots of copies of things like them around and therefore they have to work out it's you and not me.

这太精彩了。

That's brilliant.

我曾经也想过这一点。

I've thought of that.

我从未想到，意识或自我意识在规划和存在于世界中的真正价值，就在于让你能与类似你的其他实体互动。

I never thought of that that the purpose of the the really usefulness of consciousness or self awareness in the context of planning existing in the world is so you can operate with other things like you.

而且，这种‘类似你’的实体不一定是人类，也可能是其他类似的生物，确实如此。

And like you could it doesn't have to necessarily be human, it could be other kind of similar creatures, some Absolutely.

我们不是把很多自己的特质投射到宠物身上了吗？

Well, we've imbued a lot of our attributes into our pets, don't we?

或者我们试图让机器人变得像人一样。

Or we try to make our robots humanoid.

我认为这背后有深刻的原因：如果你能做出一个简化的假设——即‘你就是我，只是轮到你说话了’，那么理解这个世界就会容易得多。

I think there's a deep reason for that, that it's just much easier to read the world if you can make the simplifying assumption that basically you're me and it's just your turn to talk.

我的意思是，当我们谈论规划时，当你具体谈到规划，规划的最高体现或实现就是我们现在正在做的事情。

I mean, when we talk about planning, when you talk specifically about planning, the highest manifestation or realization of that planning is what we're doing now.

我的意思是，人类的存在状态没有比这更高的了——讨论存在的哲学和进行对话。

I mean, the human condition doesn't get any higher than this talking about the philosophy of existence and the conversation.

但在这种对话中，有一种美妙的轮流发言和相互推断的艺术，也就是心理理论。

But in that conversation, there is a beautiful art of turn taking and mutual inference, theory of mind.

我必须知道你什么时候想听。

I have to know when you want to listen.

我必须知道你什么时候想插话。

I have to know when you want to interrupt.

我必须确保你在线。

I have to make sure that you're online.

我脑子里必须有一个关于你脑子里想法的模型。

I have to have a model in my head of your model in your head.

这才是最高级、最复杂的生成模型，其中生成模型实际上拥有另一个生成模型的模型。

That's the highest and most sophisticated form of generative model, where the generative model actually has a generative model of somebody else's generative model.

我认为，我们目前正在做的事情体现了能够支持自我意识的生成模型的特性，因为如果没有这种能力，我们就会互相打断，或者像合唱团一样一起唱歌，你知道，这可能并不是我想表达的绝佳比喻，但无论如何，我们不会拥有这样的对话。

I think that and what we are doing now evinces the kinds of generative models that would support self awareness because without that, we'd both be talking over each other or we'd be singing together in a choir, you know, which is probably not that's not a brilliant analogy for what I'm trying to say, but, you know, we wouldn't have this discourse.

是的。

Yeah.

不会是这样。

Wouldn't this.

没错。

That's right.

当我打断时，你必须有所察觉。

You'd have to have as I interrupt.

我的意思是，这说得太精彩了。

I mean, that's beautifully put.

我会反复重听这次对话。

I'll I'll relisten to this conversation many times.

这段对话中充满了诗意和数学。

There's so much poetry in this and and mathematics.

让我问一个最愚蠢，或者可能是最重要的问题，作为最后一个类型的问题。

Let me ask the silliest or perhaps the biggest question as a as a last kind of question.

我们已经讨论过存在于世，以及这些实体所遵循的优化目标。

We've talked about living in existence and the objective function under which these objects would operate.

你认为我们存在的优化目标是什么？

What do you think is the objective function of our existence?

生命的意义是什么？

What what's the meaning of life?

对你而言，你认为存在的目标、满足感的来源、意义的来源是什么？作为一团物质，作为一团物质。

What do you think is the for you, perhaps, the purpose, the source of fulfillment, the source of meaning for your existence as as one blob As one blob.

在汤里。

In the soup.

我忍不住想再次以物理学家的身份来回答这个问题。

I'm I'm tempted to answer that again as a physicist.

我真的很想回答。

I'm tempted.

我预期我的行为会带来自由能的变化。

The free energy I expect consequent upon my behavior.

因此，从技术上讲，我们可以展开一场非常有趣的对话，探讨这在寻找信息、解决关于自我身份的不确定性方面包含哪些内容。

So technically, we can get this really interesting conversation about what that comprises in terms of searching for information, resolving uncertainty about the kind of thing that I am.

我猜你想要一个更个人化、更有趣的回答，但又能与上述观点保持一致。

I suspect that you want a slightly more personal and fun answer, but which can be consistent with that.

我认为答案出人意料地简单，且回归了你童年时被灌输的信念：你对自己是什么样的生物、什么样的人的认知。所有这些自我验证、以被动和具身方式最小化自由能的行为，本质上都是在实现你对自己身份的信念。

I think it's reassuringly simple and harps back to what you were taught as a child that you have certain beliefs about the kind of creature and the kind of person you are and all that self evidencing means, all that minimizing variation free energy in an inactive and embodied way means is fulfilling the beliefs about what kind of thing you are.

当然，我们所有人从小就被赋予了这些脚本和叙事，通常以睡前故事或童话的形式呈现，比如‘我是公主，我会遇到一个野兽，他会变形，然后’

Of course, we're all given those scripts, those narratives at a very early age, usually in the form of bedtime stories or fairy stories that I'm a princess, I'm gonna meet a beast who's gonna transform and it's

所以，这些叙事无处不在，从父母、朋友到社会，都在向你灌输这些故事，而你的目标函数就是去实现它们，正是如此。

So gonna be a the narratives are all around you from your parents to the friends to the society feeds these stories, and then your objective function is to fulfill Exactly.

正如你所说，这个叙事是由你的直系家庭所塑造的，同时也受到你成长环境的文化影响。

That narrative that has been encultured by your immediate family, as you say also, the of culture in which you grow up.

而你会为自己构建——再次强调，通过这种主动推断、这种被动的自我验证——你不仅在建模你的环境、你的生态位、你外部的状态，还在不断主动地改变它们，同时它们也在反过来影响你，我们正在共同完成这一切。

And you create for yourself, I mean, again, of this active inference, this inactive aspect of self evidencing, you know, not only am I modeling my environment, my econiche, my external states out there, but I'm actively changing them all the time and doing the same back, we're doing it together.

有一种同步性，意味着我在不同时间尺度上塑造着自己的文化。

There's a synchrony that means that I'm creating my own culture over different timescales.

所以现在对我来说，自私一点说，我被赋予了哪些剧本？

So the question now is for me, being very selfish, what scripts were I given?

基本上，那是爱因斯坦和夏洛克·福尔摩斯的混合体。

It basically was a mixture between Einstein and Shark Holmes.

我尽可能多地抽烟，尽量避免过多的人际接触，却又享受着一种幻想：自己是个以独特方式产生影响的流行科学家。

I smoke as heavily as possible, try to avoid too much interpersonal contact, yet enjoy the fantasy that you're a popular scientist who's going to make a difference in a slightly quirky way.

我就是在这样的环境中长大的。

That's where I grew up on.

我父亲是个工程师，热爱科学，他特别喜欢亚瑟·爱丁顿的《时空与引力》，那是第一本可理解的广义相对论著作。

My was an engineer and loved science and he loved things like Sir Arthur Edinburgh's space time and gravitation, which was the first understandable version of general relativity.

所以，我成长过程中听到的所有童话故事，都围绕着这些人物展开。

So all the fairy stories I was told as I was growing up were all about these characters.

我就不提霍比特人了，因为我已经被这些叙事深深影响了。

I'm keeping the hobbit out of this because I was quite fit by narratives.

但我想，这是一段探索的旅程。

But there's a journey of exploration, I suppose, sorts.

所以，是的，我长大后就成了我想象中，一个温文尔雅的夏洛克·福尔摩斯兼阿尔伯特·爱因斯坦会在我这种情况下做的事。

So yeah, I've just grown up to be what I imagine a mild mannered Sherlock Holmes slash Albert Einstein would would do in in my shoes.

你做得优雅而美好。

And you did it elegantly and beautifully.

卡尔，今天能和你交谈是我的巨大荣幸。

Carl, it's a huge honor talking today.

这很有趣。

It was fun.

非常感谢你抽出时间。

Thank you so much for your time.

不。

No.

谢谢你。

Thank you.

感谢

Appreciate

谢谢。

it.

感谢您收听与卡尔·弗里斯顿的对话，也感谢我们的赞助商Cash App。

Thank you for listening to this conversation with Carl Friston, and thank you to our presenting sponsor Cash App.

请考虑通过下载Cash App并使用代码Lex podcast来支持本播客。

Please consider supporting the podcast by downloading Cash App and using code Lex podcast.

如果您喜欢这个播客，请在YouTube上订阅，在Apple Podcast上给予五星评价，在Patreon上支持我们，或在Twitter上关注我Lex Friedman。

If you enjoy this podcast, subscribe on YouTube, review it with five stars on Apple Podcast, support on Patreon, or simply connect with me on Twitter at Lex Friedman.

现在，让我为您呈现卡尔·弗里斯顿的一些话。

And now let me leave you with some words from Carl Friston.

你的手臂之所以移动，是因为你预测它会动，而你的运动系统旨在最小化预测误差。

Your arm moves because you predict it will, and your motor system seeks to minimize prediction error.

感谢收听，希望下次再见。

Thank you for listening, and hope to see you next time.