Y Combinator Startup Podcast - World Models, Explained

Episode Date: July 17, 2026

Why do even our best AI models need tens of thousands of examples to learn skills that a human picks up in a handful of tries?Solving this problem is one of the great open challenges in modern AI. World models, which give AI an internal simulation of its environment, are one of the most promising paths forward.In this episode of Decoded, YC's Ankit Gupta and Francois Chaubard discuss the intuition and math behind world models, new research, and current applications in self-driving, robotics, and more.Full Transcript: https://ycrootaccess.substack.com/p/world-models-an-intuitive-introduction

One of the biggest open problems in AI right now is how to solve sample efficiency. That is, how do you get models to quickly learn new tasks or skills from relatively small amounts of training data? Humans do this incredibly well. We can learn new games, concepts, and skills, often after just a handful of tries. Our best models, on the other hand, often need tens of thousands of data points just to learn. So today we're going to discuss what many top researchers believe is the most promising path to closing that gap, world models. We're going to discuss the motivation and math behind world models, current applications, and why this approach might be the key to unlocking AGI. You and I have talked a lot about the various ways people are training models

and the sample efficiency of them. Why don't we start by just defining sample efficiency and how we intuitively think about it as humans? Yeah. So I think from my perspective, the two major problems that we have left to solve is intelligence per watt and intelligence per sample. Intelligence per watt is like how many valve perplexity points

we get per watt of spend. And then intelligence per sample is basically, if I have one additional sample in my data set, how much more intelligent am I getting? And so if I imagine I have a new tasks, like our AGI, for example, I think, like really, François Chalet has been on the forefront of this thinking and talking about intelligence as a rate of skill acquisition

versus skill acquisition. And that's very different. And so how fast do we get smarter with more and more samples? And these things are incredibly poor at getting smarter with fewer and fewer samples. And for context, you know, the RKGI test sets are a really good example of cases where humans are intuitively very good at them. Most humans can intuitively solve those puzzles with some amount of thinking and effort. But our current state-of-the-art AI systems, what people consider frontier intelligence,

basically can't do them. Right. I mean, we come into new problems with such inductive bias from K through 12, like all these math and school that we've had, that these models are kind of getting from the entire, compressing the entire, internet. And so when we come in, we're not coming in Tabula Rasa, just like bare bones. But even so that they have, you know, I don't know what percent of the internet you've read. I've read very little percent of the internet. But despite that, and having read the entire internet,

it still can't really do well and generalizing to these new tasks. So now let's think about this in like the extreme cases. In the extreme case where let's say we were perfectly sample efficient. You know, we were as sample efficient as possible. What would that mean in terms of a model that is taking a set of actions in the world? Well, I guess the perfect sample efficiency would be zero samples. And like, there are examples of this. And that sounds absurd to say. And the example, the hypothetical I'll give on this is,

imagine I had a perfect world model. Then I should never go to the environment to go and collect samples to train on. And, well, that can't possibly happen. No, it actually can happen. We do it all the time. It's called Newton's second law of motion. It's like Newton mechanics. Like, we basically know how to like get an object from point A to point B with a rocket quite easily just by following like Newton's laws of motion.

Yeah. Like when, when NASA plans to intercept an asteroid and is planning it, you know, years in advance and can set it off in a trajectory where it just glides to the right thing and intersects to the right point, that is an example of a perfect world model we've built where we're then just letting that world model act. And that system does not need to intelligent. collectively collect new samples from the environment to decide which direction to go next. It can already, it's already been pre-programmed and it can perfectly do it. Yeah. Can you imagine if, like, we needed to collect one million training examples of, like, us shooting spaceships to the moon to, like, know how to do it?

Right. It's like, it would be, we definitely wouldn't have the Apollo missions, right? But we do have that, that ability because the real world is differentiable, and we can do something called model protective control that we're going to talk about in a little bit. But even in our own brain, I was just a, uh, uh, you know, thinking about this on the drive up. But like, there's so many ways that, like, I can basically think about the things that you are going to say or what a VC is going to say when I'm, when I was pitching them or what customer might say. And even product being,

having taste, what is taste is like predicting that other people are going to like this thing. And so we've built this world model over years of entrepreneurship, 10 years of like getting it wrong, right? That maybe Bill Gates, Steve Jobs and Jensen have 50 years. of, you know, world modeling experience to know what people want. And basically, this is actually proven in the 1967 CogSai study by Richardson. They basically showed that if you take a cohort of three groups of people and you have one go practice layups in basketball, and they go and they shoot, they improve it for one hour,

they improved by like, I think it was like 24% or something like that. And then if you take the other one and they just blindfold them, And they imagine laying up a basketball. They improve it 23%. Interesting. Against the control. I mean, that's insane. It means that we have this crazy good world model.

And there's this neuroscientist at Stanford named Shaw Druckman, who basically is of the view that the entire point of the growing neocortex during the great cortical expansion 10 million years ago was to get better and better and better and better world modeling. And having just like my little VLA, which we'll define, of doing the next, predicting the next action is not as good as having a world model to lean on, either for training purposes or for test time adaptation. Yeah.

What fundamentally comes down to is, you know, we as humans, we think about our intuitive ability to think as coming from some implicit world model we have in our heads encoded by genetics and our ability to learn and whatever else. It seems like models can do surprisingly intelligent things, despite not having an explicit world model when it comes to natural language, when they're just talking. it seems like, you know, maybe under the hood, deep inside the weight somewhere, there's some kind of implicit understanding of the world. But there isn't an explicit representation of that. But it seems like in certain domains, especially in robotics and self-driving, as we'll talk about, that sort of breaks down. And, you know, maybe it would be helpful now to just think a little bit about and just sort of define some of the pieces of what makes it challenging in these different domains.

And then we can use that to kind of build up to why it's particularly hard in things like self-driving and robotic. to get these types of predictive models to work. Yeah, let's do it. So let's actually take a step back and just talk about control, reinforcement learning and define some common terms. So typically in, we teach a course called decision-making under uncertainty,

which is like the main reinforcement learning course at Stanford. I like to show a specific example of, let's say I have some drone and this is my poor little drone here and it has some mass, M. And we know that that gravity, G is pulling down on it. And it's currently at position T with velocity T, which we will collectively call the state. And to be really clear, this is going to be P X, P Y, P Z.

T, T, T, T, and V, X, V, Y, Z. It's like the six-dimensional state vector. Yep. And we have some thrust vector U that we can. control and we're trying to get to some point P star and V star, which is V star is typically zero. And so you have some platform that I want this thing that's drawn to land on. This is this control problem, right? And so let's say this is like, and we'll go through optical or optimal optimal. Optimal. Optimal. So how would I actually solve this? So the first thing

I need to know is my transition function. And so this is my state transition function. which is ST plus 1, given the previous. Given ST and my action, which I control, is UT. And so this is my state transition or dynamics function or a world model. This is a world model. This is like a very fundamental for context. You know, this equivalent to transition function you would think about in RL in general. Exactly.

And so, and then what I'm trying to learn is something called a policy, which is like what UT should I emit? given some ST. And so this is the ultimate question. What should I do? What action should I take given some state ST? And so the way that we'll solve this, and luckily we have a world model that is perfect.

It's just Newtonian physics. Newtonian physics. This is like Newton's second law of motion, which is F equals MA. And so we know that the position P-T plus 1 is going to equal P-T plus delta T-V-T, plus one half delta t squared. So everyone's taking high school,

high school physics. Yep. And the same thing for the velocity. Blap, blah, blah, and then my acceleration is the sum of the force, some of the forces,

which is going to be my UT. I think I divide by the mass. And G. And so that's it. And now I have my transition function. Now how do I get to a policy? and I'm going to apply something called model predictive control or real-time model predictive control, which is like the way that SpaceX lands the rocket on some platform in the ocean.

And what you're going to do is you're going to set up your loss function. You're going to minimize some over all T. You have U-T to infinity. And I'm going to minimize my P-star minus p-t plus v-star minus V-T-t, And usually you add this little lambda U-T, which is like how much energy you're exerting. And you can't have infinite thrust. So you typically will have to say U-T-U-Max thrust. Yeah.

That can be achieved. And so this is easily solvable with comics optimization. And so this is convex. This is convex. This is convex. The sum of convex functions is convex. This is a convex constraint. And so DCP discipline and convex programming means that I can put this into CVXP,

and it will just give me out my policy, which will be, the solution will be the optimal UT plus one, all the way to infinity. So we can solve this in closed form, basically. Because we have this world model of Newtonian physics, we can say at every step, exactly how this drone should fly so that it lands on the appropriate thing. Exactly. Under a set of constraints like max thrust. Exactly.

You'll run your log barrier or interior point, whatever, to some solver on this, and it will give me my optimal. And this would be literally the optimal path that this thing can take to get to this state. And that will minimize. And I can increase this if I want it to do the least energy path. And I make that zero if I want it to be the fastest. And so that's typically the way that you would do what I will call like deterministic differentiable control.

Right. And why differentiable? Because I can take the, I can form the little Grangian by taking this minus this constraint and take the gradient of it. And I can do monorobins. You use the fact that it's differentiable

to do the optimization. Exactly. If this is non-differential, you cannot do convex optimization and you cannot do SGD, even if it's non-convex, you could still solve and get a pretty good solution. as we do in deep learning. But if it's non-differential, you kind of can't.

There's nothing you can do. So, yeah, let's have an example then of how you could make this non-difrenchable. Like, well, what's a scenario? I guess even in like this drone scenario where now it becomes non-differenceable. Yeah. So I'll put this adversary named Unket.

Okay. And your job is to, you have another drone, let's say. Unkid's drone is to try to hit me and stop me from getting there. Now, from the position of your drone, you don't know what actions I'm going to take. Right. And so now, let's just call this the, this would be now we're definitely not deterministic. We're stochastic and non-differentialable.

And in this case, my state transition, what is ST plus one? It's going to be my say, I'm in now, my thrust, and what Uncett's going to do. Right. And these, it was all differentiable until this. new variable. Yeah. And I can't like back prop through your brain. Yeah.

To tell, say what you're going to do with your little drone controller. Yep. Right. It's completely, uh, non-differentialable now. And I,

I'm resorting and I have to resort to this awful area called reinforcement learning. Which is just super brutal and it's sprawling and there's so many different things. And you'll hear things like when you study initial, uh, um, reinforcement learning called value iteration or policy iteration. Um, and there's, There's DQN or deep Q learning or just Q learning.

There's actor critic. There's all this bag of stuff. All of this stuff ultimately comes down to ways to estimate to model this non-differentiable stochastic process. Exactly. Yeah. And so like that's basically the main thing is you're going to start talking about this as a model where I'm going to introduce this. SIE to say that this is going to be some model that's going to take in these things and then output this. and that we're going to train it over many,

many instantiations of this, and that's, so to get a better and better world model. And then I need to train some policy, A-T-S-T, and then typically you also need a value function. And that is the value of some state. And to discern between the value of different states.

And like, in this case, I don't know what a valid state is, but like, let's just say I was doing like the SpaceX with launching robots. And landing rockets in Florida. Let's just say that like there's different. If I have my launch pad here and I have a whole bunch of houses here, let's just say the path going from here to here. I may think that doing this and then coming across here and burning all these houses alive.

Right. Maybe not high value value. So I might say as an example, they typically call this like some kind of a cone here. And I might say like it's low value. to be here and it's a very high value to be to be in this colon or something, right? In a sense, the value gives you some expectation of future rewards, like the sum of future rewards you're getting. And so if you're in a bad space, you would set the value to zero or negative infinity or something.

Yeah. So we should introduce R, RT as well. And so typically, like, if you're playing Go or chess, like winning the game, you can say winning the game is plus one, minus one for losing, draw zero. That's what's done in AlphaGo. So in chess, we have these heuristics, like a pawn is worth one point, a rook is worth five, et cetera, et cetera. So you can already have reward is the difference in board state. And then this, yes, it would be the sum of my discount.

Should just do T of RT. Yeah. given and it's important also to use this nomenclature v pi and the reason why that's important is because what it what's actually happening here is this is the discounted reward following policy pie correct and that means that when i'm in this state i will take this action and then i'll end up in this to s c plus one and then i'll take this action and it's it and it's taking it greedy and so that's the value with respect to pie yeah and so ultimately what it comes down to is we are trying to still find a new policy pie. And along the way, we will use

gene learning models in various capacities. This is standard RL to estimate the value function given the rewards we're receiving. And then where world models come in is a way of incorporating all of those into some sort of joint modeling of the state and action distribution so that we can make more intelligent policies off of it. Right. And so your standard kind of set up for this is what I'm always trying to get to at the end of the day is some joint distribution, which would be ST plus 1, given where I'm at now, where I'm at now. And then this factorizes with chain rule simply to my Pi, my policy, AT given ST, and my world model.

And I'll give this. This is usually represented with theta. and this is my world model, which would be ST plus 1 given ST and AT. And these are typically learned separately. And like you can imagine, in fact, actually, you can actually learn this. This is a video generation model. And I have the frame ST, and I predict the next frame ST plus 1.

And then we'll get into this. Yeah, for those of us who kind of saw our diffusion model series, often people these days use video diffusion for exactly this. Yeah. And then what you can do, and this is like the in vogue thing to do since Danajar and the Dreamer paper series from V1 to V4 is do

action conditioning later, like similar to clip, where we will inject this like input head, input tail to come into the model to influence and enable the world model to have embodiment. What does that mean? It means that Not only can I predict, like, as a plant or tree growing on the side of the building, I can, like, see the world passing by, but I can actually influence it.

And I can change the world, and I can learn that with AT. And that's far fewer samples to do this post-action conditioning if I already have a really good ST to SC plus one world model. And so here you're saying, you know, what's also in vogue now is jointly training these versus separately training them. Exactly. So that is called the World Action Model, where some of the issues here is, one, there's all these training dynamics. If these things are just disparate training on different sets and things like that. The other issue is plainly obvious. What I have to do to actually do test time planning is I'll have to sample my, with Model 1, invoke theta, and then pass that sampled action into here and then roll it out to ST Plus 1. And it's very very, very, expensive and it's a very not real time to major issues and why like why can't we just scale up AlphaGo to like solve all the problems is because it's because of this property. If I have one

invocation to the model and it gives me both, here's the action I should take and here's the ST plus one that'll end up much, much cheaper and much much faster. Okay. So I think that's a really good segue. I think why don't we now motivate everything we just described through a series of increasingly complex environments. So I'll contend that I think the right set of environments for us to consider is chess, followed by Go, followed by self-driving, followed by robotics. All right. So let's go through a couple examples of problems that we want to apply, reinforce and learning

to. So chess is a pretty easy one. There's an 8x8 grid. And so typically when you approach any RL problem, you're going to look at star. And so the side. of the state, the number of states I can be in. So if I have these eight here and these eight, so it's eight, 16, 32, so it'd be 32 to the 64. Yes, quite large. Quite large. Then my transition function is stochastic and non-differential because you can, you don't know what the other player's

going to do. So if I'm like in playing chess.com at my house, I move and then something happens and it comes back and then now you moved and the board has changed. So I can't really differentiate through what the other player is doing. The carline of my action space is actually quite small. Even though there's 32 pieces and all that stuff, there's only eight possible moves in expectation that you can actually, that are legit moves. So like in any given a state, there's only eight-ish moves you could do. Let's just say in the beginning, I can move all my ponds.

I can move my horses. So that's 10th. Yeah. That's like not that much. So this is extremely small. And then my reward, we can use the heuristic-based approach, or we can just say, you know, plus one, zero or minus one if I lose, plus one if I win. And so this is very tractable. You say it's tractable even though there's a really big state space here.

Yeah. But why don't we talk about that for just a second? I think this is a really important point. And when you say it's tractable, you're specifically referring to the action space being small because it affects the kind of like combinatorial expansion here. Should we talk by that for just a second? Yeah. Or maybe we can add go and then kind of contrast the two. Yeah. So why don't we do that because I want to get to the alpha go the way that they solve this.

And you're right. So if I were to do this naively and I just took my SD plus one and I want to do look aheads, what I would do is I would take all of the actions I can take. So there's eight. So I would do action one, action two, action eight. And then each one of these I need to expand it for all possible states. And so now I need to do Carnality S, which we just said is this huge freaking number. And so I have to do that eight times. And I have to do it again. I have to do it again. So just looking forward, one move is like quite intractable. Although at the same time, you know, everyone starts at the same starting position.

And while it is a really large space, you know, there isn't an infinity number of potential. There's actually a really small number of game boards, even four moves. into the game, as opposed to a game where you could start in any permutation, for example, of an initial game state and what a few states down looks like. Yeah, so this is like definitely over done, because there's, there's, it's, it's much, much less than this in practice. Yes. But just naively, like, looking at, you know, what possible game states could be as a

rough math here. But this is roughly the idea. And then each one of these leaves, I need to invoke my value function. Right. which is the value of that state T plus one. And so I have to do that all many times. And we'll get this while off we go, but like this ends up being estimating the leaf node.

Because at the end of the day, my policy, A TST, I want to pick on the arg max of like the value of the following. The Argmax action, I guess it would be an A here. Yeah, A, exactly, yeah. The argmax over A of the state.

of the end state ST plus N, let's say. It's like that's the main goal here. And so for me to do that, I need to roll all this out, estimate the value, and then pick the best one. And so this quickly grows. However, and we'll see this about how we go, which actually has an even bigger state space. So I think it's 19 by 19. I don't know.

I think it's spot right now. So yeah, there's 19 by 19 grid. you can in each one it can be black, white, or nothing there. So I have three. So let's do our star again. So the cardinality of the state, I think, is going to be S two or three, my turnernerate thing here. 19 squared, I guess. It's like 3.61.

Something like that. 3601. My transition, same issue. I don't know. My action space is going to be. 361, let's say. So it's a good amount bigger than chess. Much bigger.

But it's still not enormous. Yeah. As we'll see in a second. Yeah. And so basically what they do, they call this Z, which is kind of annoying, but let's call R. And it's the terminal. It's the terminal when they won the game.

And they basically, you know, you have your trajectory, which is S-0-A-0-R-0. then all the way to the end of the game. Yep. S-N-A-N-R-N. And if you won, then all of these, all the moves that black won, all the moves that Black did get plus, all the moves that White did were minus one. And they just, that's how they create their,

their rollouts. Rollout refers to a taking end steps of play of all players, one after another of moves under a specific policy at the particular instantiation of it. Right. So let's just say, probably under this policy P, theta, T. And we're going to overload T, but like,

this is that instantiation. We froze that model. We froze that model and we play, I think it's like 70 games. And we like treat all of those. And we're going to subsample a bunch of these state action results, state action results to train our, to update our policy.

in our world model, a transition model. And what it's actually doing is we take it an ST, we give it to some theta, and it wants to output the probability of ST plus 1 being played, which is our transition function, and the value of the current state. And how do we get the value? And so the value of the current state,

well, both of them are coming out of, of the model, but basically the loss function, L theta, is going to equal, and it's going to be nearly close to this control problem one, is we have some V theta minus this Z, which we'll just call it R here, squared, and then plus, actually, so it's minus this pi, which I'll explain in a second, log p theta and I think they everyone includes this but they include it in the paper so included there as well which is the weight decay yep and so so this is basically what our loss function is then we'll play a bunch of these games and let's try to be a little bit organized here and and so this is our setup this architecture and now the most once we train this thing we do an

insane, insanely expensive task of test time planning. And so this trend in RL is this called test time planning. And a specific algorithm they use here for this is Monte Carlo research. MCTS. And so this is one of the possible things that you could do. It ends up working extremely well if you have small action spaces. Yeah. So let's let's like very intuitively talk about what MCTS does. A lot of people have heard about Monte Carlo research because AlphaGo was such a, you know, big moment. Yeah. How exactly does that map into our star and value function and policy? Yep.

So I'll take this ST. This will give me, uh, 361, uh, numbers that summed to one. And so I'll have some probability of, uh, of where these things are going to go for the, of where my, my opponent will play. Um, here. So these are like the sets of actions. Yeah. So I'm here.

so that I have all my ST plus ones. I'll have 361 of these things. And then... To be clear, this is like Action 1, Action 2, all the way to Action 361. Exactly, yeah. And the... We have to estimate the value of each one of these.

And so then we have to invoke the model all 361 times to give me values for each one of these things. And then I will select... I'll select it based on the UCB, the upper confidence bound. which is this equation that is roughly something like balancing my value function of ST plus 1, which they're going to, in the literature it would be called the Q value, because it's actually the difference between a value function and Q value is just that I have the

action as well. Yep. So it's be ST, then AT. So we'll just call that Q value, which is my, exploitation term, and then my exploration term will be something like this funky square root of N. So it's the argmax of a of my Q, and then I have this, which is the probability of this, this move being played, which we have from here of S, let's just call it ST, plus one. And then I have this term, which is this sum over n sb divided by n S.A. And what's the intuition me on this term?

So these ends is the visit count during my MCTS process. So this whole tree, I'm going to. So this tree could get really big, right? It's 361 per thing. And it's depth of 30. So you can't visit every single. whole leak though. Exactly. And so you want to keep track of which state did you end up in and what action did you take

when you were in that state. And you want to make sure that you have good exploration, right? And so the way you keep track, the way you ensure that you have good exploration is you want to not just be greedy and always pick the highest value one, because that could be local, very myopic. And so what you'll do is during this MCT process, you'll start this dictionary, which will be all zeros, of the visit count of being in this state and taking this action. And then once you go through your first rollout, you'll go here, all these things will be in it to zero. You'll have some probability.

We're going to bias it towards the higher probability of places to go. And then we'll expand those trees. And then we will update the counts that we visited this. and that will basically reduce the amount of probability that we're going to select it again, because this will reduce my exploration term. And if it's highly valued, then we're going to increase the cue on this, because this is the expected value of going down this, this, this, this. The gist of it is fundamentally like you want to take the optimal ish path,

but have enough exploration in this really expensive step you're doing here. so that you are making sure you're getting a decent chunk of the other potential leaf nodes you could traverse two in these 30-step rollouts. And so I'm going to do this MCTS simulation 800 times here. And then for all 800, I have to go through this whole process and I have to invoke the model like at least 30 times to get through all here. And so that's, you know, 27,000. 800 times 30. Yeah. So 24,000 invocations of the model to develop this tree.

And then once I have it... That's per step. Per step. Just to do one action into the game. A lot of people don't understand it. This is like, you don't, like, store this MCTS tree. You, like, you throw it away after you make the move. But once...

It's very expensive to develop this MCTS tree. And once you have it, the probabilities of traversal are actually extremely useful for training. And then you end up... Biasing it and you train it with the MCTS tree, which is like a little bit seems like circular motion or something like that like But you end up treating that as as the pie that you'll train in your loss function So you we have the R of did we win or lose we have the the pie of of what was the end result of this whole Expensive process and then at test time we are going to do these 24,000 steps every single every single move to pick the argmax that satisfies both exploration and exploration and exploitation.

In this case, this still feels somewhat tractable though because the action space is small enough where this like kind of works. Exactly. But now like let's say hypothetically maybe we can draw like an imaginary go, a game of go where it's like, you know, let's say this game of go was like a thousand. by a thousand. And so now you have A equals, you know, more or less a million. And now this, this tree we're drawing here that has to take here, this has cardinal or like, you know, width, I guess, one million. Right. And there's like S zero through S one million. And the number of,

you know, steps you would have to take here, presumably will have to be way more than 800 in order to get any reasonable kind of sampling of this. And so you're probably multiplying the test time cost of doing a rollout or of doing a next step prediction astronomically. If the game was even, let's say, you know, this is only 100 X bigger than the current game, we're not even 50x bigger than the current game. Everyone was very excited about AlphaGo. And at the time in, what was this, 2017, 2016, everyone was very excited about this.

And the important thing to pick up is that we did 800 MCTS simulations and to cover 361 possible actions on average. So that gives us about two samples, roughly on an expectation, for every single action. So here you need like two million of them for a similar depth. For a similar depth. And then that's still, to do a depth of 30, I would still have to do this times 30. This would be 60 million invocations of the model. So that better be a small model, right? that's a lot.

So yeah. That's to do a single action, to be clear. Yeah. So exactly, to do one action. So just imagine, so why AlphaGo doesn't scale. Yeah. To me, there's one, the carnality of the action space must be extremely small. If it's big, sad.

Yeah. Two, I need a perfect deterministic environment, right? like this this this this doesn't change the rules of this game don't change but like the rules of the stock market change all the time yeah the rules to like venture change all the time like the real world changes quite often so uh like uh homeless skadastistic uh and real time if you saw the movie the documentary is such such an amazing documentary i'd highly recommend it to anyone that was really good i watched it um the guy is sitting there for like 60 seconds maybe five minutes waiting for the computer to like decide. And it's kind of like, imagine that we were driving a car.

And like you took like 60 seconds to like turn the steering wheel. Everyone's dead. Like the whole car is dead. And so like, you know, now let's talk about robotics and self-driving car. And why this, why that approach kind of can scale. Yeah. I think the really good contrast here, because intuitively, I think in thinking through this exact star layout, it actually really changed how I think. about the kind of problem space of both of these two. So like, let's take self-driving car as an example. This is one, you know, many people have started

to experience for the first time because we have some self-driving cars that actually work. You have Waymo and Tesla and steel and whatnot. That seemed like they kind of work. So like let's maybe apply your same star framing here. I would contend that the state space of self-driving car is enormous

and it's actually not intuitive to me whether it's more or less large than this one, right? I mean, in a sense, the chess and Alpha ghost state space is already like more than a number of atoms in the universe or something to that effect. Right. But like just to emphasize that here, you know, you are considering, you know, surroundings, vehicle state. Yep. Like, you know, camera, like weather. Mm-hmm.

I guess the point is like, it's like massive. This is massive. This is massive. For all of the purpose is infinite. It is infinite. Yeah. And so is the space of pixels. Like what can I put in an image?

I can take an image of anything. Yes, true. And so we're able to handle it. And the same thing here, where we compress from the board state. We don't represent the board state. We compress it with a com net. So they have some deep com net that actually takes this state and converts it into a latent. And that latent compression is sufficient to kind of like do pattern matching, do

do some type of like symmetric, symmetric equivariance kind of things. And same thing with this. And even better with JPA, which we can talk about at the end there, which is like basically taking some type of state space and doing all of our optimization in the latent space, which stable diffusion did that worked extremely well, which reduces our state space dramatically because I'm in some latent high dimensional space. So like the key thing there is that, yeah, despite this state space being effectively infinite, we've actually gotten really good at compressing this.

Yeah. And we'll talk more about some of the tricks for how we actually do this in practice here, but the TLDR is, you know, where there's like 10 years of deep learning work that basically makes us extremely good at compressing that very fast. Exactly right. Exactly right. T seems to have a similar problem as before. Right.

In fact, maybe even more extreme. There's like infinity other variables around you. Right. In some ways, you'd think that it's, this is physics. Newton's laws of motion should apply. If I turn the steering wheel like this, so I hit the gas, I should be able to really easily model this. But what is non-differential is that I have, if I'm going into a circle, right,

it's like the most, the biggest issue that, that we faced in when I was doing self-driving car is like, you were imposing your will onto, maybe driving in India. I think it's going to be in a similar, right? You're imposing your will onto the environment. And like, people just kind of adapt naturally. Like, if you were doing it in a lot of motion, you were going to collide. Yes. And so that the optimal policy, if you were doing, being straight.

Newtonians here would be like, don't move. Because anything you do, you're going to crash. But it's not true. Like that, then we wouldn't function. Like cars wouldn't go down the road. And so you have to model the environment. You have to include other people in the environment and understand the embodiment of like how your action will change other people's actions.

YC's next batch is now taking applications. Got a startup in you. Apply at Ycombinator.com slash apply. It's never too early. And filling out the app will live. level up your idea. Okay, back to the video. Now let's talk about the action space. You know, like one way to look at the action space is that it seems relatively small. It seems like, well, you know, you turn the stealing wheel left or right. You hit the brake. You hit the,

you hit the gas. It doesn't seem that big. But like, how big is it actually? Like, how do we actually represent these action spaces when it comes to a realistic self-driving car scenario? Yeah, I don't know how they, how they do this nowadays. They're doing a whole bunch of like birds-eye view, different things like that. That's considered even just like a very simplified case. What do you have? You have a steering wheel that you can turn left right. You have a brake pad. Yeah. And you have the gas. And so this thing is like 365 degrees. Yeah. So it's like a one to 365, let's say, or zero to 365. Yep. And let's just say you break this up into 10 different

severities. You're already, even with just this oversimplified model, your action space cardinality. Right. It's 365,000. So that's like a 100x bigger than Alphabet. In fact, it's about the size of the example. It's a decent amount smaller than the size we said. Exactly. Break and CPS. And so, yeah, so 36,000 action space is very large.

And then even worse, unless you're Tesla, we have a bunch of video of people driving cars. We don't have video of like dash cams and like that. Like you actually don't have, again, only Tesla has this of the action as well. And so the things that you have access to that your trajectories are just like as. ST, ST plus 1, ST plus 2. So you're saying there's a decent number of these that's from like dash cam footage on YouTube or something, but not really that many either.

Yeah. Relative. And so if you wanted to do a self-driving car and you didn't want to go spend a million dollars, trillion dollars on going collecting all this data, then you want to leverage this data somehow. And this is going to be really applicable for robotics because we have a lot of videos of people doing things. Yeah.

Right? Especially with ecocentric. Like we have those videos, but what we don't have is the actions they take. Yeah. Yeah. So this is like, this is a sequence of what you're showing here. Unless you're Tesla. Unless you're Tesla.

And Tesla has this. So this is a huge competitive mode of like, what do people do in that state? And so you can behavior clone to go from here to here, from here to here, go here to here, etc. But even then, it's still very, very difficult. You have to, it's not sufficient. People think that like, okay, I have this. we have a self-driving car, right? I mean, the amount of work that they're doing at FSD is like incredible, and it's not generally available. Like, you know, it's not Waymo level yet. Would this be a good moment to briefly talk about model-free versus model-based RL? Yeah. I think that's an important distinction that's going to be relevant. We talked about more world models.

Yeah, so this is a perfect point. So model free just means that my policy pie of A-T-given ST, I have no world model involved. It's literally doing what I said. I grab a bunch of these and I go from S-to-A. Just predict the next day. That's it. And this is logic called DLA. You know, this is like giving us pretty good results.

It's behavior cloning. It's all the stuff that it's not getting us to Rosie the Roezy, the Roe's. about just yet. But in many ways, it's the closest thing that just looks like the next token prediction from LLMs that seems to scale pretty well with natural language. I mean, it's not exactly the same thing because there's no action exactly, but picking a token is not exactly the same thing, but it's very analogous to that. I basically take away the tokenizer head and I give it in an action space and I collect

a bunch of teleops data, you know, like this as the self-driving car does in Tesla. and I just taken the state, which is some image or maybe sequence of images, and then I'll output some action. And that's it. And this is, let's say, model free because I don't have a model for the environment. And then now, if I do model-based RL, I have not just some pi, but I have also my size as well here. And so by including this, I can have a much stronger policy, but it would take a lot more time to perform inference because I have to do this full test time planning. Just to remind us, that SAI is referring to this specific transition function, right? It's referring to this.

You're saying this is specifically referring to a function of ST plus 1 given ST and Action T. Yes. So it's like your ability to predict the next step. state you'll be in is the crux of it. Yep. As opposed to just directly predicting the actions. Yeah. And the main thing that I believe is that this is required for AGI.

This is what the human brain is at least in the way the human brain. Yeah. And let me go further in saying that like if you look at the billions of years of evolution, basically there's this thing called 10 million years ago called the great cortical expansion, which you see the size of a brain just explode, get bigger, bigger, bigger, bigger. exponentially up until us and it basically stops. And if the entire point of the neocortex is world modeling, what happened is we started from VLAs, this would be like ants and fish or whatever.

And fish, yeah, right? Just like very like, you know, lizard brain, whatever I call it. And then we develop this neocortex to like, you know, go from our motor cortex to actually simulate what's going to happen. And that makes us just so much smarter. And then once we get those samples, we can compress it when we sleep or otherwise. with this hippocampal, shortwave ripple ripple, whatever you want to call it,

and then that helps us develop a better policy and that marriage between the two is not only helps us train on hallucinated examples, but it also allows us to test time plan. Right. I guess the kind of extreme case then of self-driving car is kind of general robotics. Yes.

Right? So if you're like a humanoid company, like figure or pie or whatever, again, same ST-A-R, setup. I guess the gist of it is that A is now even bigger. Yeah. Right? It is like, I guess a very simple robot would be, yeah, how would you, how would you parameter at the action space? Let's take a very basic one. If I take like my six axis arm, yeah, as your standard here that we're actually working on right now in Stanford

Robotics Center, you have two degrees of freedom, two degrees of freedom, two degrees of freedom, and then you have another two for the end effector. Right. And so that's a simple end effector, not even like a, not even like a fancy hand. It's literally a one axis like, like, You know, you can rotate, but you have the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the, the one axis you me style, uh, thing. Uh, so this is eight. So, you have 16th, uh, the, uh, it's like, the, the six to the, it's like, yeah, it's an insane. Um, and so much bigger than self-driving car. Um, and even worse, like, getting teleops data is extremely painful and expensive. It's not just like, oh, we'll just get some people. Um, it's not just get some people. Um, it's just get some, um, people in the Philippines will give them like some, you know, things or whatever. It's like totally, totally doesn't work. And nor is there yet something like Tesla's fleet where there are cars deployed that people are just using and they're not even necessarily realizing that every time they turn the steering

wheel, they're providing this, this dataset for Tesla. And then even worse, you have this like what's called cross embodiment gap. And so if I were to like train this policy on Tesla Model X and I were to like, put it on a Tesla model 3, it wouldn't work. No. Like, it totally wouldn't work. Like all the, so much, so much of this, uh, the way that if, if I were to break on a model 3 versus a model X, a model X, it weighs more.

It has different dynamics, aerodynamics and things like that. And so what's actually going to happen is very different. Like the degradation you have across cross, cross embodiments is very, very strong. And clearly Tesla's figured various ways to get around that. I mean, they have these that roll up. But actually, even with Tesla's new FSD today, They don't roll out in all the cars at the same time, probably for more or less, that reason. And in this case, it's even harder now.

I mean, you have bigger differences between embodiments than a Model 3 versus Y. Yeah. And you have way bigger action spaces. You have to sum on a model. Yeah. Lane McIntosh, I played hockey with at Stanford, who now runs Tesla FSD. I can ask him, but I would bet money that they shard the data per model, per car type. Yeah, wouldn't be surprised.

I just, because that's what I would do. There's no way that, like, you know, I would try. you know, data that was collected on a Model X on a Model 3. I just wouldn't know way I would trust it. Okay. So now that we understand the basic setup here and why the action space problem is so big, why don't we talk a little bit about how world models actually fit into this? You know, maybe first, you know, I guess what didn't work about the naive world models and how

do we fix those? And then let's kind of talk about some of the newest world modeling techniques. Cool. So like in robotics in particular, it's very hard to get these, this kind of trajectories that you want, that you kind of need to train for your VLAs. and people spend up, you know, with a whole bunch of teleops data, it's very expensive, very expensive. Ideally, what we would do is take, like, data like this from someone who is just like, puts a camera on them and just like making sushi.

Okay, like, I want to make a sushi robot. How do I do it? Give it to all the sushi chefs. Don't put anything in their hands and just have them start cutting up sushi and making sushi. And ideally, we would train it in that way you were describing of, like, somehow we would train a model just on these two and then later add this afterwards. And so the first real purpose. person that, you know, went after this was Yergen-Smithuber.

Please, so he doesn't yell at us. We have to make sure we cite him. But he has his really cool paper called World Models, very aptly named. And it's basically he took these like open-a-I gym classic games, car racing and I think Doom as well. And then just like trained a model at that time was like an RNN. He had some funky Zero Order stuff. stuff in there, whatever. But basically the key premise was, I can take an environment, I can extract a whole bunch of this type of data off of it. I think he actually does actually this data, but we'll get into Dreamer where he does it in this way. And then trains a policy on only the synthetic data, the imagined rollouts. And it actually performs well in the

environment. This is the first time in my understanding that that actually happened and it actually works really well. So the key thing there, you can basically use this, if you have some predictive model of this in that case and eventually of this, you can use that as basically a synthetic training set to train your policy model and then basically fine-tuned it on real data later. Exactly. And which is just like a really powerful idea, especially since in robotics, the limiting step is access to large amounts of state action data. And so now the Dreamer series, Basically, this publishes in May of 2018. Danajar Hafner publishes Dreamer 1, I think, in November of 2018.

And then now he's been on this rampage for the last seven years publishing these papers. And Dreamer v4, I think, is the capstone of it where he basically does the same thing. And he focuses on Minecraft. And he trains a world model on this type of data. and then injects action conditioning on a very small amount of data to get to this type of world model that has the action conditioning as well.

And then samples a lot from it and then trains a policy on those synthetic, imaginated rollouts. And the policy is so good that it's the first paper to mine diamonds in Minecraft. I'm not a good Minecraft player,

but apparently that's extremely difficult. That's like next level difficulty. And it did it all on synthetic data. which is kind of crazy. And the key on luck there, yeah, use synthetic data specifically on a model trained on just this sort of state transition type of thing. Yes. And this ends up being very convenient because it turns out we as a society have a lot of this.

Exactly. Yeah. All of you too, right? He does do a very small amount of data to enable the action conditioning and that allows you to do this full simulated rollout. But yeah, it's true. So we have YouTube. We have like Flickr.

Flickr, we have all these data sets online of like people doing things. We'd like to use it. And no one has really gotten that to work. And then now that with this, these like video generated generation models, we can take that data, create a world model out of it, add action conditioning, post-train it with action conditioning for some new task that is we want it to do, chopping down wood or, you know, making sushi or folding my bed or whatever it is, only a few amount of examples.

and then we can train a policy in this neural simulation. Yeah. And we put out a video about diffusion models very recently in flow matching. I imagine that now ties very closely to this, right? Ultimately, the kind of current state-of-the-art best way to do this on basically infinity data that we have available and can keep generating is using data-of-the-art video diffusion slash flow matching models. Exactly.

Yeah. So like if you have your C-Dance or your SORA or one or whatever. Exactly. All those models, like basically the idea is now we have them and they're already trained and they're great. Let's do a small amount of action conditioning on them to get to this world model and then we can sample from it a bunch and then train. And this is exactly what Wave did with Gaia and Gaia. I think they raised $1.5 billion to basically run with this idea for self-driving car. I think a bunch of companies, Nvidia, this paper here.

is basically talking about doing exactly the same, this dream zero for robotics. What I thought was really cool about this paper is that they do exactly this process where they have this joint model of state transitions and actions. They train it by first instantiating it with the open source one video diffusion model. And then it only takes them about 500 hours of teleop data, which is basically exactly this, to get it to be pretty good and they have a lot of clever tricks that allowed it to be cross-embodiment and working on scene tasks with relatively small amounts of data.

And it really is taking basically the exact concept, I believe, from the Dreamer paper and applying it specifically to these robot embodiments. Exactly. And it turns out it actually works actually better than I would have anticipated it working. Yeah, so I think that this is basically the path to, it was the path, I believe,

the path to get humans to be as good as we are. Genetically, over the last 10, 20 million years of evolution, A bigger world model helps for training and for test time planning. And I think it'll be the same thing as true as for robotics. What's also cool is there's a bunch of applications of this, the things outside of robotics too. I mean, there was a weather planning paper, for example. There was Gencast paper,

which I think applies a relatively similar concept in terms of how they model, you know, literally the world's weather with something like this. Yeah, we have to talk about the world model for the world. Yeah, so basically they do this exact same thing where, you know, the key unlocks for this whole thing was getting diffusion to work in very high dimensional state spaces like we talked about in the last lecture and then learning to use that to action condition in the way that

he's done. But they did this for the entire world with this exact same diffusion steps, which go from some, and they go back to two time steps, lag of order to AR-T, two for the set of decisions there. And I basically predict the next state of the world based on those things with this Lingyevin diffusion rollouts. My big assertion is that it was necessary for the human brain to develop world modeling. I actually just saw this paper that I wanted to make sure to call out because that was so

great out of University of Washington where they say explicitly in the abstract, each cortical area estimates both latent sensory states and actions and the cortex as a whole predicts the consequences of those actions. That sounds like a world model to me. Yeah. Right. It's actually describing exactly these two equations here. Exactly.

Right. Where we're estimating both the sensory latent states and actions. I mean, I guess it's really the joint model that we showed earlier. Right. What he's describing here. It's exactly this equation. Yep. Yeah.

Exactly right. And so if it works in us, it should work. in robotics. And I think that that takes us the rest of the distance. Why don't we talk briefly about latent world models, especially the JEPA concept, because I think there's been a number of papers that use JEPA as an element of their, I guess, architecture. Why don't we just briefly introduce JEPA and how it fits into the current landscape of world modeling? Yeah. In classic RL, you'll have, like, you know, if you study Q learning, for example, you basically keep this matrix called the Q matrix. Yep. And it's going to be S by A. And so I have this S by A, states and actions.

And each one, I need some amount of counts of being in this state action. And I take the average value of taking that action in this state. And that's my cue value there. And it's a little bit more complicated in that. There's a bell in the equation. I'll just back up all this stuff like that. But so this scales horribly because as the cardinality in my space, space gets bigger and my kind of action space gets bigger stuff I don't have enough to

I become less and less sample efficient right in case of like robots or whatever state is like yeah it's this whole thing we described earlier right it's absolutely massive because it has all of these elements in it couldn't really enumerate a huge group and so the classic trick I mean since I took you know see it 229 with and wrong in 2012 is you do this stick a neural Lepwork on it exactly and you basically are just going to compress that state into some lower dimensional state space this is actually predates deep learning. We were doing stuff like this. I think my first paper was basically doing something like this, basically turning like a grid into like a bunch of like pyramids.

And like and the state was how much I'm in pyramid one or pyramid two or whatever. But anyway, the neural networking can just do this. And so basically what, uh, the key idea in JIPA, if I have, um, an image one and I have image two and I have image three. I can do my world modeling, my world modeling of ST plus one, given ST and AT, in pixel space and have, this is, let's say, at time, T, T plus one, T plus two, et, et et cetera, and I have to actually predict now the full image. That's extremely expensive from a computation standpoint and also from like a sample efficiency standpoint. What I can do instead is put this through some comnet. Some encoder.

Some encoder. And then I'll get a latent for T. And I'll have a latent for T plus one. And I'll have a latent for ZT plus two. And then I'll have from this, from ZT, I want to predict Z T plus one hat. And my goal is to make this and this. make my loss function will be something very simple like want to minimize this that's it now this now this doesn't work this collapses hard and so what happens like it's basically just if you out if it just predict zero yeah yeah just output zeros which the model will learn to do and I'm actually incorporating this into my current research right now and so what you need to do is something called sigreg or uh this is one technique vic reg is another where basically I add this another

term that basically says, uh, I want the, um, over a large enough batch size, I want the, the distribution of Z, T plus one to follow a Gaussian, you know, it's kind of like a normalize it, like a like a batch norm type of track. I mean, not in the same case. And if, if it's zero, it can't be this. Yeah. Right. Cause then this is non zero. And so maybe I think that there's probably this or something like that. But basically this prevents it from modal collapse. And it makes it do something good. And this is the most recent paper for the audience is L.EWM, Ellie World Model, which is super, super great. However, to be completely frank, this is self-supervised learning, super great. It doesn't work that well. If you were to not do these techniques,

and there's a bunch of other techniques that you can do, it will actually outperform much better that are, let's say, for example, if I'm going to do an LLM and you have like, you know, Francois likes sushi, which is definitely true. And I tokenize this into a bunch of different tokens here. And this is token ID 6, 19, 28, whatever. And I look up the encoding into this. And that's going to be E1, E2, E3, etc. what you can actually do is have the LLM output,

the LM will take in these things and will output the next token. And so it would be like, let's call it H. This would be the low jets coming out of it, T plus 1. And what you can do is actually have this be close to E, T plus 1. And a lot of people are playing with this idea. and getting rid of the cross-send should be lost entirely. And so if you were to do this,

it actually is a proxy for the cross-dentropy loss, and there is no cross-tentropy loss. And the cross-tentropy head is actually very expensive. And so this is very cheap. And like this is literally just grabbing it. So people are playing around with this idea and basically as a cheaper proxy for the cross-tentropy loss. So there's lots of different ideas on basically taking this J-Pai idea

to not just pixels, but to L.S. LNs as well. Yeah, interesting, yeah. So just to define what JPA is, it's joint embedding predictive architecture. I think one of the things I find cool about this JPA idea is it feels like an idea we see over and over in deep learning. There's a version of this idea that's basically the staple diffusion idea. Yeah.

There's a version of this idea that in my company training, graph convolutional neural networks to design drugs we use to do, you know, latent variable generation, for example. And it's like, it's an idea that comes back over and over and then has this, yeah, various tricks that it actually takes to get it to work in practice. Okay, now we have a pretty good sense for how world models work. We have a pretty good sense for what the state of the art looks like. If we trust this paper, and it seems like these kind of work on robots, too, I mean, this paper is only from the end of last year, this year.

And it seems like they have various methods that allow you to train on relatively small amounts of data that's tractable and pre-trained on the diffusion models. So are we good? We're done. Does it all work? Yeah. This is 2016. 26 will be the year of the robot.

We're going to have Rose of the Robot in your house. Yeah, no. I don't think so. What are one or two? Because there's lots of open problems remaining. What are like a few open problems? Maybe we can emphasize here that the community can go emphasize working on. Yeah.

So I think the first one is that PINS doesn't really work. What is PINs? Physics-informed neural networks. So Pins doesn't really work. There's physics-informed neural networks. And so basically if almost all of the self-driving car data looks like this, the car is driving down the road. And let's just say, for example, I have, you know, a house here.

And I want to train the model on not driving into the house. And so let's say I put it into a state right here to drive into the house. What's going to happen is because almost all the data is, is like looks like this, trying down the road. This will just turn magically into like a highway. And I'm just like, boo, just don't worry. You can crash all. It basically needs like a ton of data not to do that either from simulation for that to not happen.

In fact, I actually don't even know if because of the data distribution, there's no data here. There's almost all the data here. And like when you're training a neural network, it has a tendency to collapse. If you don't keep the mini batch composition, like very even over the, you know, over the class space or whatever you want to want to call it. But like you'd have to train on, you have to be very careful about your data mixing to make sure you get this right to solve this problem

that no one really has. But even then, if you take just a simple thing like this, this is like the, the conic example, and I have some sign wave. And I want, and I have these as my X,

and I have these as my Y. So this is complete interpolation. No. That may mess this up, but why? Like this. We can't get to like machine precision. We can't, what is it? What is it?

What is it? What is it? What is it? We can't, we, the SGD will not get to zero, effectively zero. So we'll always have some residual. And for us to be like a really good world model to simulate body interactions, like to simulate this. What's going to happen when I do this?

And like, let's say that I'm trying to be LeBron James. Like, there's like, I saw this one. video of um step curry tribling about a basketball on a court and he just felt that there was a dead spot in the court and he because he's so good and he knows exactly the physics of what's going to happen if i hit this you know the ball with this force like the ball's going to come back exactly the spot and it just didn't and he knew it wasn't him it was the the court and he found a dead spot in the court like that's how good the the human brain is at world modeling in my opinion i think it's an sgd issue i think it's probably an architecture issue i think sam altman just kind of came

and just said that he thinks that there's definitely an architecture that's going to be more performing than the transformer. I think he's right. I think the transformer doesn't do compression in the time domain at all. It just keeps around everything. So anyway, so I think that the getting higher fidelity in the world model is extremely important. One, I think two. Seems like test time probably is going to be a big thing, like adaptation.

Exactly. Test time planning, how quickly the human brain can, you know, in times in sports and things like that, when you're playing tennis, I think you're a tennis player, like how quickly we can adapt to what a player is doing and things like that. We're not going to sleep and like retraining. We're very quick to adapt to a new environment. It's like the out of distribution prediction. Exactly. And like one little data point, we can like quickly adapt to that new thing and change. I think there's been a lot of papers.

on like basically estimating the friction coefficients. And so like those can change over time if you go to a human environment or not. For example, like this friction might change. And that's important in control. And so you need to estimate that very quickly and adapt. And that these models just kind of don't have a mechanism to do it. Yeah. And then I guess there's like the practical speed elements of these, right?

A lot of these are doing some sort of expensive planning step. And we're doing some sort of like, we're kind of hacking around it with this retraining process and synthetic data. But even so, like, to really get maximum performance right now, you'd want to do something that's closer to like the AlphaGo style rollout. And that's extremely slow. Right. The MCTS process can't happen.

The other thing that is pretty crazy about the way that the brain works is that, like, everything is kind of running autonomously. And so like, you might be like in the middle of saying sentence one and be like, oh, actually, no, something else. And so like, what does happen there? It's like type 1 and type 2 thinking are happening at the same time in some way. And so like there's definitely, you know, some really cool mix of these like heterogeneous models. And like some are overriding others and like taking control of the motor cortex and like commanding the body to do a thing.

Okay. But on the flip side now, we talked in the past video about the squint test and how we felt that auto-aggressive LLMs maybe don't pass the squint test. Why don't we reintroduce what the squint test was for a second? And then maybe let's think about whether this passes the squint test, despite all those limitations. Yeah. And the swin test for me, I think, is like, this comes from the Jan Lacoon. We didn't need flapping wings to achieve flight. And to that, I say, well, we did need two wings.

And like, if I squint and I look at a bird and I squint and I look at a plane, I'm like, yeah. It's kind of similar. It looks right. Similarly, if I squint and I look at the human brain and I squint and I look at all these world models, we have like this VLA, this action policy, and that they're doing test time planning together and things like that. It's getting really close. It's much, much closer.

It seems closer than an auto-aggressive L-LM. And that's like this concept of a world model of, you know, implicitly predicting future states and actions feels intuitively like what our brain doing. And it seems like there's some neuroscience evidence to support that. I mean, I'm getting to the conclusion that I think that the brain is the optimizer, not the model. And that the brain is,

emits, like has models that it invokes, but the brain is somehow also the optimizer itself. And so in that way, it doesn't pass the squint because, like, you know, something magical is happening when you're sleeping. There's no intelligent species that we're aware of that have any amount of intelligence that don't sleep. And so like octopuses, dolphins, all those stuff, elephants, they all sleep. There's some reason for that. And that seems like a really thing about like the evolutionary recourse of sleeping. Like you get eaten when you sleep. So like for the benefit of sleeping should be so so much better to outperform that. So I think we don't have this idea of awake sleep in our current architecture, but I can imagine I'm like simulating, you know,

compressed from the hippocampus some like experience in the day. I'm like training on more of those examples, right? You're like collecting a whole bunch of these experience rollouts and then you're updating your policy function. There's got to be something like there's this thing called shortwave ripple where like the hippocampus when you're sleeping like emits these, uh, spike trains that are actually reversed from when they actually happen back in through the both both the hemispheres and for like seven times and then it like stops so like there's something happening there that's very training something yeah and if you don't sleep then you don't up you don't have long-term memory right right and so like there's definitely a reason why we're we're training things that happened uh into our brain so where does that put us now that we have all this work happening with world models how should we think about what's coming ahead in these next few years in the

research community. Yeah, I think that like we're going to see a lot more of these world models in robotic policies. I think that's going to unlock probably full solve driving would be like one of those examples that they can get the real timeness of it. They can probably, they can probably solve it with more compute to like have parallel things. And you probably don't need it for like most standard things. Maybe like, you know, getting out of weird parking jams and like things like that would take us sometime. Similar to the rose of the robot, which we've always wanted to have a Rose of the Robot to, like, you know, clean up my room for me. I think that, like, this feels like we're getting to good enough that we can pay up for data and

compute to get to Rose of the Robot. It does feel like that. It'll be expensive to collect the data and do the Dreamer sequence of going from state to state and then getting the action conditioning to work. But, like, I feel like it should work. Yeah. I mean, it's pretty cool is we see a lot of companies at YC working at every step of this from the collecting egocentric data, collecting the teleop data, training their own world models

and action models, building new embodiments, and then making ways of adapting those embodiments. And it feels like this is the first year where you see demos where you're like, okay, this actually like kind of is starting to look like it's going somewhere. Yeah. And it seems like a very exciting year. Yeah. So anyway, I think that there are real AI problems to solve still. We talked about pins.

We talked about the real time issues. And then on the robotic side, there's real issue. Like, it's amazing how effective our epidermis is in terms of we can detect tactile. Oh, epitermis. Yeah, epidermis. Our tactile, we can detect sheer force. We can detect temperature. And it's everywhere.

And so, like, versus, you know, like the, we get like one little sensor that only does tactile. We don't have the friction component. We don't have temperature. We don't have all these, the feeling. We can't estimate coefficient of friction very quickly. I can touch something to say, oh, this is smooth, this is rough. We don't have any of that. And if I numb your hands, I actually had this experience just recently.

If I numb your hands, like, you actually can't tie your shoes. So you can't perform control. And so, like, yeah, if you, like, you know, if you train enough, on enough human data tying your laces, do I think you can do it with no feedback? Maybe. Maybe. But, like, how much would you need if you did actually have the human, like, touch? Like, I think it would be so much easier. Yeah.

Well, there's a lot of more research to do then. Yeah, yeah. Rezwe, thanks so much for joining us. Thanks so much for watching, everyone. We'll be back for the next episode of Decoded.