#86 – David Silver: AlphaGo, AlphaZero, and Deep Reinforcement Learning

David Silver leads the reinforcement learning research group at DeepMind and was lead researcher on AlphaGo, AlphaZero and co-lead on AlphaStar, and MuZero and lot of important work in reinforcement learning. Support this podcast by signing up with these sponsors: – MasterClass: https://masterclass.com/lex – Cash App – use code “LexPodcast” and download: – Cash App (App Store): https://apple.co/2sPrUHe – Cash App (Google Play): https://bit.ly/2MlvP5w EPISODE LINKS: Reinforcement learning (book): https://amzn.to/2Jwp5zG This conversation is part of the Artificial Intelligence podcast. If you would like to get more information about this podcast go to https://lexfridman.com/ai or connect with @lexfridman on Twitter https://twitter.com/lexfridman, LinkedIn https://www.linkedin.com/in/lexfridman/, Facebook https://www.facebook.com/lexfridman, Medium https://medium.com/@lexfridman, or YouTube https://www.youtube.com/lexfridman where you can watch the video versions of these conversations. If you enjoy the podcast, please rate it 5 stars on Apple Podcasts https://podcasts.apple.com/us/podcast/artificial-intelligence/id1434243584, follow on Spotify https://open.spotify.com/show/2MAi0BvDc6GTFvKFPXnkCL, or support it on Patreon https://www.patreon.com/lexfridman. Here’s the outline of the episode. On some podcast players you should be able to click the timestamp to jump to that time. OUTLINE: 00:00 – Introduction 04:09 – First program 11:11 – AlphaGo 21:42 – Rule of the game of Go 25:37 – Reinforcement learning: personal journey 30:15 – What is reinforcement learning? 43:51 – AlphaGo (continued) 53:40 – Supervised learning and self play in AlphaGo 1:06:12 – Lee Sedol retirement from Go play 1:08:57 – Garry Kasparov 1:14:10 – Alpha Zero and self play 1:31:29 – Creativity in AlphaZero 1:35:21 – AlphaZero applications 1:37:59 – Reward functions 1:40:51 – Meaning of life

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Lex Fridman?

  The following is a conversation with David Silver, who leads the Reinforcement Learning Research Group at DeepMind, and was the lead researcher on AlphaGo, AlphaZero, and co-led the AlphaStar and MuZero efforts, and a lot of important work in reinforcement learning in general. I believe AlphaZero is one of the most important accomplishments in the history of artificial intelligence. And David is one of the key humans who brought AlphaZero to life, together with a lot of other great researchers at DeepMind. He's humble, kind, and brilliant. We were both jet lagged, but didn't care and made it happen. It was a pleasure and truly an honor to talk with David.

• 0:43Lex Fridman?

  This conversation was recorded before the outbreak of the pandemic. For everyone feeling the medical, psychological, and financial burden of this crisis, I'm sending love your way. Stay strong, we're in this together, we'll beat this thing. This is the Artificial Intelligence Podcast. If you enjoy it, subscribe on YouTube, review it with five stars on Apple Podcasts, support it on Patreon, or simply connect with me on Twitter at @LexFriedman, spelled F-R-I-D-M-A-N. As usual, I'll do a few minutes of ads now, and never any ads in the middle that can break the flow of the conversation. I hope that works for you and doesn't hurt the listening experience. Quick summary of the ads. Two sponsors, MasterClass and Cash App.

• 1:27Lex Fridman?

  Please consider supporting the podcast by signing up to MasterClass at masterclass.com/lex and downloading Cash App and using code LEXPODCAST. This show is presented by Cash App, the number one finance app in the App Store. When you get it, use code LEXPODCAST. Cash App lets you send money to friends, buy Bitcoin, and invest in the stock market with as little as $1. Since Cash App allows you to buy Bitcoin, let me mention that cryptocurrency in the context of the history of money is fascinating. I recommend A Scent of Money as a great book on this history. Debits and credits on ledgers started around 30,000 years ago. The US dollar created over 200 years ago.

• 2:12Lex Fridman?

  And Bitcoin, the first decentralized cryptocurrency, released just over 10 years ago. So, given that history, cryptocurrency is still very much in its early days of development. But it's still aiming to, and just might redefine, the nature of money.

• 2:29Lex Fridman?

  So again, if you get Cash App from the App Store or Google Play and use the code LEXSPodcast, you get $10 and Cash App will also donate $10 to FIRST, an organization that is helping to advance robotics and STEM education for young people around the world. This show is sponsored by MasterClass. Sign up at masterclass.com/lex to get a discount and to support this podcast. In fact, for a limited time now, if you sign up for an All-Access Pass for a year, you get to get another All-Access Pass to share with a friend. Buy one, get one free. When I first heard about MasterClass, I thought it was too good to be true. For $180 a year, you get an All-Access Pass to watch courses from, to list some of my favorites.

• 3:12Lex Fridman?

  Chris Hadfield on space exploration, Neil deGrasse Tyson on scientific thinking and communication, Will Wright, the creator of SimCity, and Sims on game design, Jane Goodall on conservation, Carlos Santana on guitar. His song Europa could be the most beautiful guitar song ever written. Garry Kasparov on chess, Daniel Negrono on poker, and many, many more. Chris Hadfield explaining how rockets work and the experience of being launched into space alone is worth the money. For me, the key is to not be overwhelmed by the abundance of choice. Pick three courses you want to complete, watch each of them all the way through. It's not that long, but it's an experience that will stick with you for a long time, I promise. It's easily worth the money.

• 3:56Lex Fridman?

  You can watch it on basically any device. Once again, sign up on masterclass.com/lex to get a discount and to support this podcast. And now, here's my conversation with David Silver.

• 4:26Lex Fridman?

  What was the first program you ever written? And what programming language? Do you remember?

• 4:31Speaker 01

  I remember very clearly, yeah. My, my parents brought home this BBC Model B microcomputer. It was just this fascinating thing to me. I was about seven years old and couldn't resist just playing around with it. So I think first program ever was writing my name out in different colors and getting it to loop and repeat that. And there was something magical about that, which just led to more and more.

• 5:01Lex Fridman?

  How did you think about computers back then? Like the, the magical aspect of it, that you can write a program and there's this thing that you just gave birth to that's able to create sort of visual elements and live in its own? Or did you not think of it in those romantic notions? Was it more like, oh, that's cool. I can, I can solve some puzzles.

• 5:22Speaker 01

  It was always more than solving puzzles. It was something where, you know, there was this limitless possibilities. Once you have a computer in front of you, you can do anything with it. That's, I used to play with Lego with the same feeling. You can make anything you want out of Lego, but even more so with a computer. You know, you don't, you're not constrained by the amount of kit you've got. And so I was fascinated by it and started pulling out the, you know, the user guide and the advanced user guide and then learning. So I started in basic and then, you know, later six, five, oh, two. My father was also became interested in the, in this machine and gave up his career to go back to school and, and study for a, a master's degree in, in artificial intelligence. Funny enough, at Essex University when I was, when I was seven.

• 6:05Speaker 01

  So I was exposed to those things at an early age. He showed me how to program in Prolog and do things like querying your family tree. And those are some of my early, earliest memories of trying to, trying to figure things out on a computer.

• 6:21Lex Fridman?

  Those are the early steps in computer science programming. But when did you first fall in love with artificial intelligence or with the ideas, the dreams of AI?

• 6:31Speaker 01

  I think it was really when I when I went to study at university. So I was an undergrad at Cambridge and studying computer science. And and I really started to question, you know, what really are the goals? What's the goal? Where do we want to go with with computer science? And it seemed to me that the the only step of major significance to take was to try and recreate something akin to human intelligence. If we could do that, that would be a major leap forward. And that idea certainly wasn't the first to have it, but it nestled within me somewhere and and became like a bug. You know, I really wanted to to crack that problem.

• 7:15Lex Fridman?

  So you thought it was like you had a notion that this is something that human beings can do, that it is possible to create an intelligent machine?

• 7:24Speaker 01

  Well, I mean, unless you believe in something metaphysical, then what are our brains doing? Well, at some level, they're information processing systems, which are

• 7:37Speaker 01

  able to take whatever information is in there, transform it through some form of program and produce some kind of output, which enables that that human being to do all the amazing things that they can do in this incredible world.

• 7:48Lex Fridman?

  So so then, do you remember the first time you've written a program that because you also had an interest in games? Do you remember the first time you were in a program that beat you in a game?

• 8:02Lex Fridman?

  That more beat you at anything, sort of achieved super David Silver level performance?

• 8:11Speaker 01

  So I used to work in the games industry. So for five years, I I programmed games for my first job. So it was an amazing opportunity to get involved in a startup company. And so I I was involved in in building AI at that time.

• 8:29Speaker 01

  And so for sure, there was a sense of building handcrafted, what people used to call AI in the games industry, which I think is not really what we might think of as as AI in its fullest sense, but something which is able to to.

• 8:45Speaker 01

  Take actions in a way which makes things interesting and challenging for the for the for the human player. And at that time, I was able to build, you know, these handcrafted agents, which in certain limited cases could do things which which were able to do better than than me, but mostly in these kind of Twitch-like scenarios where where they were able to do things faster or because they had some pattern which was able to exploit repeatedly. I think if we're talking about real AI, the first experience for me came after that, when I I realized that this path I was on wasn't taking me towards, it wasn't, it wasn't dealing with that bug, which I still had inside me to really understand intelligence

• 9:29Speaker 01

  and try and and try and solve it. Everything people were doing in games was, you know, short-term fixes rather than long-term vision. And so I went back to study for my PhD, which was, funnily enough, trying to apply reinforcement learning to the game of Go. And I built my first Go program using reinforcement learning, a system which would, by trial and error, play against itself and was able to learn which patterns were actually helpful to predict whether it was going to win or lose the game and then, you know, choose the moves that led to the combination of patterns that would mean that you're more likely to win. And that system, that system beat me.

• 10:08Lex Fridman?

  And how did that make you feel?

• 10:10Speaker 01

  Made me feel good.

• 10:12Lex Fridman?

  I mean, was there sort of the, yeah, is it's a mix of a sort of excitement and was there a tinge of sort of like, almost like a fearful awe? You know, it's like in Space, 2001 Space Odyssey, kind of realizing that you've created something that that is, you know, that is, that's achieved human-level intelligence in this one particular little task. And in that case, I suppose neural networks weren't involved.

• 10:41Speaker 01

  There were no neural networks in those days. This was pre-deep learning revolution. But it was a principled self-learning system, based on a lot of the principles which which people are still using in in deep reinforcement learning. How did I feel? I, I think I found it immensely satisfying that a system which was able to learn from first principles for itself was able to reach the point that it was understanding this domain.

• 11:13Speaker 01

  Better than better than I could and able to outwit me. I I don't think it was a sense of awe. It was a sense that satisfaction that this that something I felt should work had worked.

• 11:25Lex Fridman?

  So to me, AlphaGo, and I don't know how else to put it, but to me, AlphaGo and AlphaGo Zero, mastering the game of Go is, again, to me, the most profound and inspiring moment in the history of artificial intelligence. So you're one of the key people behind this achievement. And I'm Russian, so I really felt the first sort of seminal achievement when Deep Blue beat Garry Kasparov in 1987. So as far as I know, the AI community at that point largely saw the game of Go as unbeatable in AI, using the the sort of the state of the art brute force methods, search methods.

• 12:05Lex Fridman?

  Even if you consider, at least the way I saw it, even if you consider arbitrary exponential scaling of compute, Go would still not be solvable, hence why it was thought to be impossible. So, given that the game of Go was impossible to to master, what was the dream for you? You just mentioned your PhD thesis of building the system that plays Go. What was the dream for you that you could actually build a computer program that achieves world class, not necessarily beats the world champion, but achieves that kind of level of playing Go?

• 12:41Speaker 01

  First of all, thank you. That's very kind words. And funnily enough, I just came from a panel where I was actually in a conversation with Garry Kasparov and Murray Campbell, who was the author of Deep Blue. And it was their first meeting together since the since the match. So that just occurred yesterday. So I'm literally fresh from that experience. So these are amazing moments when they happen. But where did it all start? Well, for me, it started when I became fascinated in the game of Go. So Go, for me, I've I've grown up playing games. I've always had a fascination in in in board games. I played chess as a kid. I played Scrabble as a kid. When I was at university, I discovered the game of Go.

• 13:25Speaker 01

  And and to me, it just blew all of those other games out of the water. It was just so deep and profound in its in its complexity with endless levels to it. What I discovered was that I could devote endless hours to this game.

• 13:42Speaker 01

  And I knew in my heart of hearts that no matter how many hours I would devote to it, I would never become a, you know, a grand master. Or there was another path, and the other path was to try and understand how you could get some other intelligence to play this, this game better than I would be able to. And so even in those days, I, I had this idea that, you know, what if, what if it was possible to, to build a program that could crack this? And as I started to explore the domain, I discovered that, you know, this was really the, the domain where people felt deeply that if progress could be made in Go, it would really mean a giant leap forward for AI. It was the, the challenge where all other approaches had failed.

• 14:27Speaker 01

  You know, this is coming out of the, the era you mentioned, which was in some sense, the, the golden era for, for the classical methods of AI, like heuristic search. In the nineties, you know, they all, they all fell one after another, not just chess with Deep Blue, but checkers, backgammon, Othello. There were numerous cases where, where systems built on top of heuristic search methods with, you know, these high performance systems had been able to defeat the human world champion in each of those domains. And yet, in that same time period, there was a million dollar prize available for

• 15:06Speaker 01

  the game of Go, for the first system to beat a human professional player. And at the end of that time period, at year 2000, when the prize expired, the strongest Go program in the world was defeated by a nine-year-old child, when that nine-year-old child was giving nine free moves to the computer at the start of the game to try and even things up. Yeah. And computer Go expert beat that strongest, same strongest program with 29 handicap stones, 29 free moves. So that's what the state of affairs was when I became interested in this problem in around 2000 and 2003, when I, I start, started working on computer Go. There was nothing, there was just, there was very, very little in the way of progress

• 15:51Speaker 01

  towards meaningful performance, again, at anything approaching human level. And so people. They it wasn't through lack of effort. People have tried many, many things. And so there was a strong sense that that something different would be required for Go than than had been needed for all of these other domains where AI had been successful. And maybe the single clearest example is that that Go, unlike those other domains, had this kind of intuitive property that a Go player would look at a position and say, hey, you know, here's this mess of black and white stones. But from this mess, oh, I can I can predict that that this part of the board has become my territory, this part of the board has become your territory.

• 16:34Speaker 01

  And I've got this overall sense that I'm going to win and that this is about the right move to play. And that intuitive sense of judgment of being able to evaluate what's going on in a position.

• 16:45Speaker 01

  It was pivotal to humans being able to play this game and something that people had no idea how to put into computers. So this question of how to evaluate in a position, how to come up with these intuitive judgments, was the key reason why Go was so hard. In addition to its enormous search space, and the reason why methods which had succeeded so well elsewhere failed in Go. And so people really felt deep down that that, you know, in order to crack Go, we would need to get something akin to human intuition. And if we got something akin to human intuition, we'd be able to solve, you know, much, many, many more problems in AI. So for me, that was the moment where it's like, okay, this is not just about playing the game of Go. This is about something profound.

• 17:30Speaker 01

  And it was back to that bug which had been itching me all those years. Now, this is the opportunity to do something meaningful and transformative and and I guess a dream was born.

• 17:40Lex Fridman?

  That's a really interesting way to put it. So almost this realization that you need to find, formulate Go as a kind of a prediction problem versus a search problem was the intuition. I mean, maybe that's the wrong crude term, but the.

• 17:56Lex Fridman?

  To give it us the ability to kind of intuit things about positional structure of the board. Now, okay, but what about the learning part of it? Did you have a sense that you have to, that learning has to be part of the system? Again, something that hasn't really, as far as I think, except with TD Gammon in the 90s with RL a little bit, hasn't been part of those state of the art game playing systems.

• 18:25Speaker 01

  So I strongly felt that learning would be necessary. And that's why my PhD topic back then was trying to apply reinforcement learning to the game of Go. And not just learning of any type, but I felt that the only way to really have a system to progress beyond human levels of performance wouldn't just be to mimic how humans do it, but to understand for themselves. And how else can a machine hope to understand what's going on, except through learning? If you're not learning, what else are you doing? Well, you're putting all the knowledge into the system. And that just feels like a, something which decades of AI have told us is maybe not a dead end, but certainly has a ceiling

• 19:09Speaker 01

  to the capabilities. It's known as the, you know, knowledge acquisition bottleneck, that the more you try to put into something, the more brittle the system becomes. And so you just have to have learning. You have to have learning. That's the only way you're going to be able to get a system which has sufficient knowledge in it, you know, millions and millions of pieces of knowledge, billions, trillions of a form that it can actually apply for itself and understand how those billions and trillions of pieces of knowledge can be leveraged in a way which will actually lead it towards its goal without conflict or other issues.

• 19:44Lex Fridman?

  Yeah, I mean, if I put myself back in that time, I just wouldn't think like that. Without a good demonstration of RL, I would, I would think more in the symbolic AI, like the Not learning, but sort of a simulation of a knowledge base, like a growing knowledge base, but it would still be sort of pattern based. Lot like basically have little rules that you kind of assemble together into a large knowledge base.

• 20:13Speaker 01

  Well, in a sense, that was the state of the art back then. So if you look at the Go programs, which had been competing for this prize I mentioned, they were an assembly of different specialized systems, some of which used huge amounts of human knowledge to describe how you should play the opening, how you should, all the different patterns that were required to to play well in the game of Go. End game theory, combinatorial game theory, and combined with more principled, search-based methods, which were trying to solve for particular subparts of the game, like life and death, connecting groups together. All these amazing sub-problems that just emerge in the game of Go,

• 20:57Speaker 01

  there were there were different pieces all put together into this collage, which together would try and play against a human. And although not all of the pieces were handcrafted, the overall effect was nevertheless still brittle, and it was hard to make all these pieces work well together. And so really, what I was pressing for and the main innovation of the approach I took was to go back to first principles and say, "Well, let's let's back off that and try and find a principled approach where the system can learn for itself." Just from the outcome, like, you know, learn for itself. If you try something, did that did that help or did it not help?

• 21:39Speaker 01

  And only through that procedure can you arrive at knowledge which is which is verified. The system has to verify it for itself, not relying on any other third party to say this is right or this is wrong. And so that principle was already, you know, very important in those days. But unfortunately, we were missing some important pieces back then.

• 22:00Lex Fridman?

  So before we dive into maybe discussing the beauty of reinforcement learning, let's take a step back, we kind of skipped skipped it a bit, but the the rules of the game of Go, what's the the elements of it perhaps contrasting to chess that sort of you really enjoy it as a human being and also that make it really difficult as a AI machine learning problem?

• 22:30Speaker 01

  So the game of Go has remarkably simple rules. In fact, so simple that people have speculated that if we were to meet alien life at some point, that we wouldn't be able to communicate with them, but we would be able to play a game of Go with them. They probably have discovered the same rule set. So the game is played on a on a 19 by 19 grid and you play on the intersections of the grid and the players take turns. And the aim of the game is very simple. It's to surround as much territory as you can, as many of these intersections with your stones and to surround more than your opponent does. And the only nuance to the game is that if you fully surround your opponent's piece, then you get to capture it and remove it from the board and it counts as your own territory. Now, from those very simple rules, immense complexity arises. There's kind

• 23:15Speaker 01

  of profound strategies in how to surround territory, how to kind of trade off between making solid territory yourself now compared to building up influence that will help you acquire territory later in the game, how to connect groups together, how to keep your own groups alive, which which patterns of stones are most useful compared to others. There's just immense knowledge and human Go players have played this game for, it was discovered thousands of years ago, and human Go players have built up this immense knowledge base over the years. It's studied very deeply and played by something like 50 million players across the

• 23:55Speaker 01

  world, mostly in China, Japan and Korea, where it's an important part of the culture, so much so that it's considered one of the four ancient arts that was required by Chinese scholars. So there's a deep history there.

• 24:08Lex Fridman?

  But there's interesting qualities. So if I had to compare to chess, chess is in the same way as it is in Chinese culture for Go and chess in Russia is also considered one of the sacred arts. So if we contrast sort of Go with chess, there's interesting qualities about Go.

• 24:26Lex Fridman?

  Maybe you can correct me if I'm wrong, but the the evaluation of a particular static board is not as reliable. Like, you can't in chess, you can kind of assign points to the different units.

• 24:40Speaker 01

  Yeah.

• 24:40Lex Fridman?

  And it's kind of a pretty good measure of who's winning, who's losing. It's not so clear.

• 24:46Speaker 01

  Yeah. So in the game of Go, you know, you find yourself in a situation where both players have played the same number of stones, actually captures a strong level of play happen very rarely, which means that at any moment in the game, you've got the same number of white stones and black stones. And the only thing which differentiates how well you're doing is this intuitive sense of, you know, where are the territories ultimately going to form on this board? And when you if you look at the complexity of a real Go position, you know, it's it's it's mind boggling that that kind of question of what will happen in in 300 moves from now, when you when you see just a scattering of 20 white and black stones intermingled. And and so that that challenge is the reason why position evaluation

• 25:31Speaker 01

  is so hard in Go compared to to other games. In addition to that, has an enormous search space. So there's around 10 to the 170 positions in the game of Go. That's an astronomical number. And that search space is is so great that traditional heuristic search methods that were so successful and things like Deep Blue and and chess programs just kind of fall over and go.

• 25:53Lex Fridman?

  So at which point did reinforcement learning enter your life, your research life, your way of thinking? We just talked about learning, but reinforcement learning is a very particular kind of learning, one that's both philosophically sort of profound.

• 26:09Speaker 01

  Yeah.

• 26:10Lex Fridman?

  But also one that's pretty difficult to get to work as if we look back in the earth, at least the early days. So when did that enter your life? And how did that work progress?

• 26:19Speaker 01

  So I had just finished working in the games industry, this startup company. And I took I took a year out to discover for myself exactly which path I wanted to take. I knew I wanted to study intelligence, but I wasn't sure what that meant at that stage. I really didn't feel I had the tools to decide on exactly which path I wanted to follow. So during that year, I I read a lot. And one of the things I read was Sutton and Bato, the sort of seminal textbook on an introduction to reinforcement learning. And When I read that textbook, I I just had this resonating feeling that this is what I understood

• 27:02Speaker 01

  intelligence to be. And this was the path that I felt would be necessary to go down to make progress in in AI. So I got in touch with Rich Sutton and asked him if he would be interested in supervising me on a PhD thesis in in ComputerGo. And he he basically said that if he's still alive, he'd be happy to. But unfortunately, he'd been, you know, struggling with very serious cancer for some years, and he really wasn't confident at that stage that he'd even be around to see the

• 27:42Speaker 01

  end of it. But fortunately, that part of the story worked out very happily. And I found myself out there in Alberta. They've got a great games group out there with a history of fantastic work in in board games as well, as Rich Sutton, the father of RL. So it was the the natural place for me to go in some sense to to study this question. And the more I looked into it, the more the more strongly I I felt that this wasn't just the path to progress in ComputerGo, but really, you know, this this was the thing I'd been looking for. This was

• 28:20Speaker 01

  really an opportunity to to frame what intelligence means, like what is what are the goals of AI in a clear, single, clear problem definition, such that if we're able to solve that clear, single problem definition, in some sense, we've we've cracked the problem of AI.

• 28:38Lex Fridman?

  So to you, reinforcement learning ideas, at least sort of echoes of it, would be at the core of intelligence. It is at the core of intelligence. And if we ever create a human-level intelligence system, it would be at the core of that kind of system.

• 28:54Speaker 01

  Let me say it this way, that I think I think it's helpful to separate out the problem from the solution. So I see the problem of intelligence. I would say it can be formalized as the reinforcement learning problem, and that that formalization is enough to capture.

• 29:11Speaker 01

  Most, if not all, of the things that we mean by intelligence, that they can all be brought within this framework and gives us a way to access them in a meaningful way that allows us as scientists to understand intelligence and us as computer scientists to build them. And so in that sense, I feel that it gives us a path, maybe not the only path, but a path towards AI. And so do I think that any system in the future that's, you know, solved AI would have to have RL within it? Well, I think if you ask that, you're asking about the solution methods. I would say that if we have such a thing, it would be a solution to the RL problem.

• 29:54Speaker 01

  Now, what particular methods have been used to get there? Well, we should keep an open mind about the best approaches to actually solve any problem. And, you know, the things we have right now for reinforcement learning, maybe, I believe they've got a lot of legs, but maybe we're missing some things. Maybe there's going to be better ideas. I think we should keep a, you know, let's remain modest and we're at the early days of this field and there are many amazing discoveries ahead of us.

• 30:22Lex Fridman?

  For sure. The specifics, especially of the different kinds of RL approaches currently, there could be other things that fall under the very large umbrella of RL. But if it's, if it's okay, can we take a step back and kind of ask the basic question of what is, to you, reinforcement learning?

• 30:38Speaker 01

  So reinforcement learning is the study and the science and the problem of intelligence in the form of an agent that interacts with an environment. So the problem you're trying to solve is represented by some environment, like the world in which that agent is situated. And the goal of RL is clear, that the agent gets to take actions. Those actions have some effect on the environment and the environment gives back an observation to the agent saying, you know, this is what you see or sense. And one special thing which it gives back is called the reward signal, how well it's doing in the environment. And the reinforcement learning problem is to simply take actions over time so as to maximize that reward signal.

• 31:24Lex Fridman?

  So a couple of basic questions. What types of RL approaches are there? So I don't know if there's a nice, brief, in words way to paint the picture of sort of value-based, model-based, policy-based reinforcement learning.

• 31:42Speaker 01

  Yeah. So now if we think about, okay, so there's this ambitious problem definition of RL. It's really, you know, it's truly ambitious. It's trying to capture and encircle all of the things in which an agent interacts with an environment and say, well, how can we formalize and understand what it means to to crack that? Now let's think about the solution method. Well, how do you solve a really hard problem like that? Well, one approach you can take is is to decompose that that very hard problem into into pieces that work together to solve that hard problem. And and so you can kind of look at the decomposition that's inside the agent's head, if you like, and ask, well, what form does that decomposition take? And some of the most common pieces that people use when they're kind of putting this system,

• 32:24Speaker 01

  the solution method together, some of the most common pieces that people use are whether or not that solution has a value function. That means is it trying to predict, explicitly trying to predict how much reward it will get in the future? Does it have a a representation of a policy? That means something which is deciding how to pick actions. Is is that decision-making process explicitly represented? And is there a model in the system? Is there something which is explicitly trying to predict what will happen in the environment? And so those three pieces are, to me, some of the most common building blocks. And I understand the different choices in RL as choices of whether or not to use those building blocks when you're trying to decompose the the solution.

• 33:09Speaker 01

  You know, should I have a value function represented? Should I have a policy represented? Should I have a model represented? And there are combinations of those pieces and, of course, other things that you could add into the picture as well. But those those three fundamental choices give rise to some of the branches of RL with which we're very familiar.

• 33:25Lex Fridman?

  And so those, as you mentioned, there is the choice of what's specified or modeled explicitly. And the idea is that that all of these are somehow implicitly learned within the system. So it's almost a choice of. How you approach a problem, do you see those as fundamental differences or are these almost like um small specifics, like the details of how you solve the problem, but they're not fundamentally different from each other?

• 33:57Speaker 01

  I think the the fundamental idea is is maybe at the higher level, the fundamental idea is um the first step of the decomposition is really to say, well, how are we really going to solve any kind of problem where you're trying to figure out how to take actions? And just from this stream of observations, you know, you've got some agent situated in its sensory motor stream and getting all these observations in, getting to take these actions, and what should it do? How can you even broach that problem? You know, maybe the complexity of the world is so great that you can't even imagine how to build a system that would that would understand how to deal with that. And so the first step of this decomposition is to say, well, you have to learn, the system has to learn for itself. And so note that

• 34:39Speaker 01

  the reinforcement learning problem doesn't actually stipulate that you have to learn. Like, you could maximize your awards without learning, it would just wouldn't do a very good job of it.

• 34:48Lex Fridman?

  Yes.

• 34:49Speaker 01

  So learning is required because it's the only way to achieve good performance in any sufficiently large and complex environment. So so that's the first step. And so that step gives commonality to all of the other pieces. Because now you might ask, well, what should you be learning? What does learning even mean? You know, in in this sense, you know, learning might mean, well, you're trying to update the parameters of some system, which is then the thing that actually picks the actions. And and and those parameters could be representing anything. They could be parameterizing a value function or a model or a policy. And so in that sense, there's a lot of commonality in that whatever is being represented there is the thing which is being learned, and it's being learned with the ultimate goal of maximizing rewards. But

• 35:35Speaker 01

  But the way in which you decompose the problem is really what gives the semantics to the whole system. Like, are you trying to learn something to predict well, like a value function or a model? Are you learning something to perform well, like a policy? And the form of that objective is kind of giving the semantics to the system. And so it really is, at the next level down, a fundamental choice. And we have to make those fundamental choices as system designers, or enable our algorithms to be able to learn how to make those choices for themselves.

• 36:06Lex Fridman?

  So then, the next step you mentioned, the very first thing you have to deal with is, can you even take in this huge stream of observations and do anything with it? So the natural next basic question is, what is deep reinforcement learning? And what is this idea of using neural networks to deal with this huge incoming stream?

• 36:31Speaker 01

  So amongst all the approaches for reinforcement learning, deep reinforcement learning is one family of solution methods that tries to utilize powerful representations that are offered by neural networks to represent any of these different components of the solution, of the agent, like whether it's the value function, or the model, or the policy. The idea of deep learning is to say, well, here's a powerful toolkit that's so powerful that it's universal, in the sense that it can represent any function, and it can learn any function. And so if we can leverage that universality, that means that

• 37:13Speaker 01

  whatever we need to represent for our policy, or for our value function, for our model, deep learning can do it. So that deep learning is one approach that offers us a toolkit that has no ceiling to its performance, that as we start to put more resources into the system, more memory and more computation, and more data, more experience, more interactions with the environment, that these are systems that can just get better and better and better at doing whatever the job is they've asked them to do, whatever we've asked that function to represent, it can...

• 37:47Speaker 01

  Learn a function that does a better and better job of representing that that that knowledge, whether that knowledge be estimating how well you're going to do in the world, the value function, whether it's going to be choosing what to do in the world, the policy, or whether it's understanding the world itself, what's going to happen next, the model.

• 38:03Lex Fridman?

  Nevertheless, the the fact that neural networks are able to learn incredibly complex representations that allow you to do the policy, the model, or the value function is, at least to my mind, exceptionally beautiful and surprising. Like, was it, is it surprising, was it surprising to you? Can you still believe it works as well as it does? Do you have good intuition about why it works at all and works as well as it does?

• 38:36Speaker 01

  I think, let me take two parts to that question. I think it's not surprising to me that the idea of reinforcement learning works, because in some sense, I think it's the, I feel it's the only thing which can ultimately. And so I feel we have to, we have to address it, and there must be success is possible, because we have examples of intelligence. And it must at some level be able to, possible to acquire experience and use that experience to do better in a way which is meaningful to environments of the complexity that humans can deal with. It must be. Am I surprised that our current

• 39:17Speaker 01

  systems can do as well as they can do? I think one of the big surprises for me and a lot of the community is really the fact that deep learning can continue to perform so well, despite the fact that these neural networks that they're representing have these incredibly nonlinear kind of bumpy surfaces, which to our kind of low dimensional intuitions make it feel like surely you're just going to get stuck and learning will get stuck because you won't be able to make any further progress. And yet. The big surprise is that learning continues and and these what appear to be

• 40:02Speaker 01

  local optima turn out not to be because in high dimensions, when we make really big neural nets, there's always a way out and there's a way to go even lower and then you're still not in a local optima because there's some other pathway that will take you out and take you lower still. And so no matter where you are, learning can can proceed and do better and better and better without bound. And so that is a surprising and beautiful property of neural nets, which I find elegant and beautiful and and somewhat shocking that it turns out to be the case.

• 40:37Lex Fridman?

  As you said, which I really like, to our low dimensional intuitions, that's surprising.

• 40:45Speaker 01

  Yeah, yeah, we're very, we're very tuned to working within a three dimensional environment. And so to start to visualize what a billion dimensional neural network surface that you're trying to optimize over, what that even looks like is very hard for us. And so I think that really, if you try to account for for the

• 41:09Speaker 01

  essentially the AI winter where where people gave up on neural networks. I think it's really down to that that lack of ability to generalize from from low dimensions to high dimensions. Because back then, we were in the low dimensional case. People could only build neural nets with, you know, fifty nodes in them or something. And to to imagine that it might be possible to build a billion dimensional neural net and it might have a completely different qualitatively different property was very hard to anticipate. And I think even now, we're starting to build the the theory to support that. And and it's incomplete at the moment, but all of the theory seems to be pointing in the direction that indeed, this is an approach which which truly is universal, both in its representational capacity, which was known,

• 41:54Speaker 01

  but also in its learning ability, which is which is surprising.

• 41:57Lex Fridman?

  And it it makes one wonder what else we're missing due to our low dimensional intuitions that that will seem obvious once it's discovered.

• 42:08Speaker 01

  I often wonder, you know, when we one day do have AIs which are superhuman in their abilities to to understand the world.

• 42:22Speaker 01

  What will they think of the algorithms that we developed back now? Will it be, you know, looking back at these days and, you know, and thinking that, will we look back and feel that these algorithms were naive first steps? Or will they still be the fundamental ideas which are used even in a hundred thousand, 10,000 years?

• 42:43Lex Fridman?

  Yeah, they'll, they'll watch back to this conversation and with a smile, maybe a little bit of a laugh. I mean, my sense is, I think, just like when we used to think that the sun revolved around the earth, they'll see our systems of today, reinforcement learning, as too complicated. That the answer was simple all along. There's something, just like you said in the game of Go. I mean, I love the systems of cellular automata, that there's simple rules from which incredible complexity emerges. So it feels like there might be some really simple approaches.

• 43:27Lex Fridman?

  Just like Richard Sutton says, right? These simple methods with compute over time seem to prove to be the most effective.

• 43:37Speaker 01

  I 100% agree. I think that if we try to anticipate what will generalize well into the future, I think it's likely to be the case that it's the simple, clear ideas which will have the longest legs and which will carry us furthest into the future. Nevertheless, we're in a situation where we need to make things work today. And sometimes that requires putting together more complex systems where we don't have the full answers yet as to what those minimal ingredients might be.

• 44:08Lex Fridman?

  So speaking of which, if we could take a step back to Go, what was Mogo and what was the key idea behind the system?

• 44:17Speaker 01

  So back during my PhD on Computer Go, around about that time, there was a major new development in, which actually happened in the context of Computer Go. And And it was really a revolution in the way that heuristic search was was done. And and the idea was essentially that

• 44:40Speaker 01

  a position could be evaluated, or a state in general, could be evaluated, not by humans saying whether that position is good or not, or even humans providing rules as to how you might evaluate it. But instead, by allowing the system to randomly play out the game until the end multiple times, and taking the average of those outcomes as the prediction of what will happen. So for example, if you're in the game of Go, the intuition is that you take a position and you get the system to kind of play random moves against itself all the way to the end of the game, and you see who wins. And if black ends up winning more of those random games than white,

• 45:22Speaker 01

  well, you say, hey, this is a position that favors white. And if white ends up winning more of those random games than black, then it favors white. So that idea was known as Monte Carlo

• 45:36Speaker 01

  search, and a particular form of Monte Carlo search that became very effective and was developed in Computer Go, first by Remy Coulomb in 2006, and then taken further by others, was something called Monte Carlo Tree Search, which basically takes that same idea and uses that that insight to evaluate every node of a search tree as evaluated by the average of the random playouts from that from that node onwards. And this idea was very powerful and suddenly led to huge leaps forward in the strength of Computer Go playing programs.

• 46:12Speaker 01

  And among those, the strongest of the Go playing programs in those days was a program called Mogo, which was the first program to actually reach human master level on small boards, nine by nine boards. And so this was a program by someone called Sylvain Jelley, who's a good colleague of mine, but I worked with him a little bit in those days, part of my PhD thesis. And Mogo was a first step towards the latest successes we saw in Computer Go. But it was still missing a key ingredient. Mogo was evaluating purely by random rollouts against itself. And in a way, it's it's truly remarkable that random play should give you anything at all.

• 46:55Lex Fridman?

  Yes.

• 46:56Speaker 01

  Like, why, why in this perfectly deterministic game that's very precise and involves these very exact sequences, why is it that that random randomization is is helpful? And so the intuition is that randomization captures something about the the nature of the of the the search tree that that from a position that you're you're understanding the nature of the search tree from that node onwards by by by using randomization. And this was a very powerful idea.

• 47:26Lex Fridman?

  And I've seen this in in other spaces. I talked to Richard Karp and so on. Randomized algorithms somehow magically are able to do exceptionally well and and simplifying the problem somehow. Makes you wonder about the fundamental nature of randomness in our universe. It seems to be a useful thing. But so from that moment, can you maybe tell the origin story and the journey of AlphaGo?

• 47:53Speaker 01

  Yeah. So programs based on Monte Carlo tree search were a first revolution in the sense that they led to suddenly programs that could play the game to any reasonable level. But they they plateaued. It seemed that no matter how much effort people put into these techniques, they couldn't exceed the level of amateur dan level Go players. So strong players, but not not anywhere near the level of of professionals, never mind the world champion. And so that brings us to the birth of AlphaGo, which happened in the context of a startup company known as DeepMind.

• 48:31Lex Fridman?

  I heard of them.

• 48:32Speaker 01

  Where a a project was born. And the project was really a scientific investigation where myself and Ajuan and an intern, Chris Madison, were exploring a scientific question. And that scientific question was really

• 48:54Speaker 01

  Is there another fundamentally different approach to this key question of Go, the key challenge of how can you build that intuition and how can you just have a system that could look at a position and understand what move to play or how well you're doing in that position, who's going to win? And so the deep learning revolution had just begun. The systems like ImageNet had suddenly been won by deep learning techniques back in 2012. And following that, it was natural to ask, well, you know, if if deep learning is able to scale up so effectively with images to to understand them enough to to classify them, well, why not go? Why?

• 49:35Speaker 01

  Why not take the black and white stones of the Go board and build a system which can understand for itself what that means in terms of what move to pick or who's going to win the game, black or white? And so that was our scientific question, which we we were probing and trying to understand. And as we started to look at it, we discovered that we could build a system. So in fact, our very first paper on AlphaGo was actually a pure deep learning system, which was trying to answer this question. And we showed that actually, a pure deep learning system with no search at all was actually able to reach human ban level, master level, at the full game of Go, 19 by 19 boards.

• 50:18Speaker 01

  And so without any search at all, suddenly we had systems which were playing at the level of the best Monte Carlo tree search systems, the ones with randomized rollouts.

• 50:28Lex Fridman?

  So first of all, sorry to interrupt, but that's kind of a groundbreaking notion. That's that's like basically a definitive step away from the a couple of decades of essentially search dominating AI.

• 50:42Speaker 01

  Yeah.

• 50:43Lex Fridman?

  So what how do them? Make you feel, was it surprising from a scientific perspective in general, how to make you feel?

• 50:50Speaker 01

  I found this to be profoundly surprising. In fact, it was so surprising that we had a bet back then. And like many good projects, you know, bets are quite motivating. And the bet was, you know, whether it was possible for a system based purely on deep learning, no search at all, to beat a down level human player. And so we had someone who joined our team, who was a down level player. He came in and we had this first match against him and.

• 51:23Lex Fridman?

  Which side of the bet were you on, by the way? Did you have the losing or the winning side?

• 51:28Speaker 01

  I tend to be an optimist with the power of deep learning and reinforcement learning. So the system won and we were able to beat this human down level player. And for me, that was the moment where it was like, okay, something, something special is afoot here. We have a system which without search is able to already just look at this position and understand things as well as a strong human player. And from that point onwards, I really felt that reaching, that reaching the top levels of human play, you know, professional level, world champion level, I felt it was actually an inevitability.

• 52:11Speaker 01

  And if it was an inevitable outcome, I was rather keen that it would be us that achieved it. So we scaled up. This was something where, you know, so I had lots of conversations back then with Demis Tsarvas, the head of DeepMind, who was extremely excited.

• 52:34Speaker 01

  And we made the decision to scale up the project, brought more people on board. And so AlphaGo became something.

• 52:44Speaker 01

  Where we had a clear goal, which was to try and crack this outstanding challenge of AI to see if we could beat the world's best players. And this led within the space of not so many months to playing against the European champion Fan Huai in a match which became memorable in history as the first time a Go program had ever beaten a professional player. And at that time, we had to make a judgment as to whether, when and whether we should go and challenge the world champion. And this was a difficult decision to make. Again, we were basing our predictions on our own progress and had to estimate based

• 53:27Speaker 01

  on the rapidity of our own progress, when we thought we would exceed the level of the human world champion. And we tried to make an estimate and set up a match, and that became the AlphaGo versus Lee Sedol match in 2016.

• 53:43Lex Fridman?

  And we should say, spoiler alert, that AlphaGo was able to defeat Lee Sedol.

• 53:50Speaker 01

  That's right, yeah.

• 53:51Lex Fridman?

  So maybe we could take even a broader view. AlphaGo involves both learning from expert games and, as far as I remember, a self-play component to where it learns by playing against itself. But in your sense, what was the role of learning from expert games there? And in terms of your self-evaluation, whether you can take on the world champion, what was the thing that you're trying to do more of, sort of train more on expert games? Or was there now another... I'm asking so many poorly phrased questions, but did you have a hope, a dream that self-play

• 54:36Lex Fridman?

  would be the key component at that moment yet?

• 54:41Speaker 01

  So in the early days of AlphaGo, we we used human data to explore the science of what deep learning can achieve. And so when we had our first paper that showed that it was possible to predict the winner of the game, that it was possible to suggest moves, that was done using human data. Oh, solely human data. Yeah. And and and so the reason that we did it that way was, at that time, we were exploring separately the deep learning aspect from the reinforcement learning aspect. That was the part which was which was new and unknown to to to to me at that time, was how far could that be stretched? Once we had that, it then became natural to try and use that same representation and see if we could learn for ourselves using

• 55:22Speaker 01

  that same representation. And so right from the beginning, actually, our goal had been to build a system using self-play. And to us, the human data, right from the beginning, was an expedient step to help us, for pragmatic reasons, to go faster towards the goals of the project than we might be able to starting solely from self-play. And so in those days, we were very aware that we were choosing to to use human data, and that might not be the long-term Holy Grail of AI, but that it was something which was extremely useful to us. It helped us to understand the system. It helped us to build deep learning representations

• 56:01Speaker 01

  which were clear and simple and easy to use. And so really, I would say it's it served a purpose, not just as part of the algorithm, but something which I continue to use in our research today, which is trying to break down a very hard challenge into pieces which are easier to understand for us as researchers and develop. So if you if you use a component based on human data, it can help you to understand the system, such that then you can build the more principled version later that that does it for itself.

• 56:32Lex Fridman?

  So as I said, the AlphaGo victory, and I don't think I'm being sort of romanticizing this notion. I think it's one of the greatest moments in the history of AI. So were you cognizant of this magnitude of the accomplishment at the time? I mean, are you cognizant of it even now? Because to me, I feel like it's something that would, we mentioned what the AGI systems of the future will look back. I think they'll look back at the AlphaGo victory as like, holy crap, they figured it out. This is where this is where it started.

• 57:08Speaker 01

  Well, thank you again. I mean, it's funny because I guess I've been working on, I've been working on computer Go for a long time. So I'd been working at the time of the AlphaGo match on computer Go for more than a decade. And throughout that decade, I'd had this dream of what would it be like to, what would it be like really to actually be able to build a system that could play against the world champion? And I imagined that that would be an interesting moment, that maybe, you know, some people might care about that and that this might be, you know, a nice achievement. But I think when I arrived in Seoul and discovered the legions of journalists that were following us around and the 100 million people that

• 57:51Speaker 01

  were watching the match online live, I realized that I'd been off in my estimation of how significant this moment was by several orders of magnitude. And so there was definitely an adjustment process to realize that this, this was something which the world really cared about and which was a watershed moment. And I think there was that moment of realization.

• 58:18Speaker 01

  But it's also a little bit scary because, you know, if you go into something thinking it's going to be maybe of interest, and then discover that 100 million people are watching, it suddenly makes you worry about whether some of the decisions you'd made were really the best ones or the wisest, or were going to lead to the best outcome. And we knew for sure that there were still imperfections in AlphaGo, which were going to be exposed to the whole world watching. And so, yeah, it was, I think, a great experience. And I feel privileged to have been part of it, privileged to have led that amazing team. I feel privileged to have been in a moment of history, like you say. But also lucky that, you know, in a sense, I was insulated from, from the knowledge of,

• 59:03Speaker 01

  I think it would have been harder to focus on the research if the full kind of reality of, of what was going to come to pass had, had been known to me and, and, and the team. I think it was, you know, we were, we were in our bubble and we were working on research and we were trying to answer the scientific questions. And then bam, you know, the, the public sees it. And, and I think it was, it was, it was better that way in retrospect.

• 59:24Lex Fridman?

  Were you confident that, I guess, what were the chances that you could get the win? So, and just like you said, I'm, I'm a little bit more familiar with another accomplishment that we may not even get a chance to talk to, I talked to Oriol Vinyalis about Alpha Star, which is another incredible accomplishment. But here, you know, with Alpha Star and beating the StarCraft, there is like already a track record with AlphaGo. This is like the really first time you get to see reinforcement learning, uh, face the best human in the world. So what was your confidence like? What was the odds?

• 1:00:01Speaker 01

  Well, we actually, um,

• 1:00:02Lex Fridman?

  Was there a bet?

• 1:00:04Speaker 01

  Funnily enough, there was. Uh, so, so just before the match, um, we, we weren't betting on anything concrete, but we all held out a hand. Everyone in the team held out a hand at the beginning of the match. Um, and the number of fingers that they had out on that hand was, um, supposed to represent how many games they thought we would win against Lee Sedol. And there was an amazing spread in the, in the team's predictions. But I have to say, I predicted 4-1. Um, and, and the reason was based purely on, on data. So I'm a scientist first and foremost. And one of the things which we had established was that AlphaGo in around one in five games would develop something which we called a delusion, which was a kind of, you know, hole in its, in its knowledge where it wasn't

• 1:00:49Speaker 01

  able to fully understand everything about the position. And that, that hole in its knowledge would persist for tens of moves throughout the game. Um, and we knew two things. We knew that if there were no delusions, that AlphaGo seemed to be playing at a level that was far beyond any human capabilities. But we also knew that If there were delusions, the opposite was true. And and in fact, you know, that's that's what came to pass. We saw we saw all of those outcomes and and Lisa Doll in in one of the games played a really beautiful sequence that that that AlphaGo just hadn't predicted. And after that, it it led it into this situation where it was unable to really understand the position fully and and and found itself in one of these these delusions.

• 1:01:34Speaker 01

  So so indeed, yeah, 4-1 was the outcome.

• 1:01:37Lex Fridman?

  So yeah, and can you maybe speak to it a little bit more? What were the the the five games like what what happened? Is there interesting things that that come to memory in terms of the play of the human or the machine?

• 1:01:50Speaker 01

  So I remember all of these games vividly, of course. You know, moments like these don't come too often in the lifetime of a of a scientist. And the the first game was was magical because it was the the first time that a a computer program had defeated a world champion in in this grand challenge of Go. And and there was a moment where where AlphaGo invaded Lisa Doll's territory towards the end of the game. And and that's quite an audacious thing to do. It's like saying, hey, you thought this was going to be your territory in the game, but I'm going to stick a stone right in the middle of it and and and prove to you that I can break it up.

• 1:02:34Speaker 01

  And Lisa Doll's face just dropped. He wasn't expecting a computer to to do something that audacious.

• 1:02:43Speaker 01

  The second game became famous for a move known as move 37. This was a a move that was played by AlphaGo that was broke all of the conventions of of Go that the Go players were so shocked by this. They they they thought that maybe the operator had made a mistake. They they thought that there was something crazy going on. And and it just broke every rule that Go players are taught from a very young age. They're just taught, you know, you this kind of move called a shoulder hit. You you you can only play it on the third line or the fourth line. And AlphaGo played it on the fifth line. And and it turned out to be a brilliant move and made this beautiful pattern in the middle of the board that ended up winning the game. And so.

• 1:03:27Speaker 01

  This really was a clear instance where we could say computers exhibited creativity, that this was really a move that was something humans hadn't known about, hadn't anticipated, and computers discovered this idea. They were the ones to say, actually, you know, here's a new idea, something new, not in the domains of human knowledge of the game. And now the humans think this is a reasonable thing to do, and it's part of Go knowledge now. The third game, something special happens when you play against a human world champion, which again, I hadn't anticipated before going there, which is, you know, these players are amazing.

• 1:04:10Speaker 01

  Lee Sedol was a true champion, 18-time world champion, and had this amazing ability to probe AlphaGo for weaknesses of any kind. And in the third game, he was losing, and we felt we were sailing comfortably to victory, but he managed to, from nothing, stir up this fight and build what's called a double co, these kind of repetitive positions. And he knew that historically, no computer Go program had ever been able to deal correctly with double co positions, and he managed to summon one out of nothing. And so for us, you know, this was a real challenge. Like, would AlphaGo be able

• 1:04:50Speaker 01

  to deal with this, or would it just kind of crumble in the face of this situation? And fortunately, it dealt with it perfectly. The fourth game was amazing in that Lee Sedol appeared to be losing this game. You know, AlphaGo thought it was winning, and then Lee Sedol did something which I think only a true world champion can do, which is, he found a brilliant sequence in the middle of the game, a brilliant sequence that led him to really just transform the position. It kind of,

• 1:05:24Speaker 01

  he found just a piece of genius, really. And after that, AlphaGo, its evaluation just tumbled. It thought it was winning this game, and all of a sudden, it tumbled and said, oh, now I've got no chance. And it started to behave rather oddly at that point. In the final game, for some reason, we as a team were convinced, having seen AlphaGo in the previous game, suffer from delusions. We, we as a team, were convinced that it was suffering from another delusion. We were convinced that it was mis-evaluating the position and that, that something was going terribly wrong. And it was only in the last few moves of the game that we realized that, actually, although it had been predicting it was going to win all the way through, it really was. And, um, and so somehow, you

• 1:06:09Speaker 01

  know, it just taught us yet again that you have to have faith in, in your systems. When they, when they exceed your own level of ability and your own judgment, you have to trust in them to, to know better than, than you, the designer. Once, um, you've, you've bestowed in them the ability to, to judge better than you can, then trust the system to do so.

• 1:06:30Lex Fridman?

  So, just like in the case of Deep Blue beating Garry Kasparov, so Garry was, is, I think, the first time he's ever lost, actually, to anybody. And, I mean, there's a similar situation with Lee Sedol. It's a, it's a tragic, uh, it's a tragic loss for humans, but a beautiful one. I think that's kind of, uh, from the tragedy sort of emerges over time, emerges a kind of inspiring story. But, uh, Lee Sedol recently announced his retirement. I don't know if we can look too deeply into it, but he did say that, even if I become

• 1:07:15Lex Fridman?

  number one, there's an entity that cannot be defeated. So, what do you think about these words? What do you think about his retirement from the game ago?

• 1:07:24Speaker 01

  Well, let me take you back, first of all, to the first part of your comment about Garry Kasparov, because actually, at the panel yesterday, um, he specifically said that when he first lost to Deep Blue, He viewed it as a failure. He viewed that this had been a failure of his. But later on in his career, he said he'd come to realize that actually it was a success. It was a success for everyone because this marked a transformational moment for AI. And so even for Garry Kasparov, he came to realize that that moment was pivotal and actually meant something much more than his personal loss in that moment.

• 1:08:06Speaker 01

  Lee Sedol, I think, was much more cognizant of that even at the time. So in his closing remarks to the match, he really felt very strongly that what had happened in the AlphaGo match was not only meaningful for AI, but for humans as well. And he felt as a Go player that it had opened his horizons and meant that he could start exploring new things. It brought his joy back for the game of Go because it had broken all of the conventions and barriers and meant that suddenly, suddenly, anything was possible again. And so, yeah, it was sad to hear that he'd retired, but he's been a great world champion over many, many years.

• 1:08:48Speaker 01

  And I think that he'll be remembered for that ever more. He'll be remembered as the last person to beat AlphaGo. I mean, after that, we increased the power of the system and the next version of AlphaGo beats the other strong human players 60 games to nil. So what a great moment for him and something to be remembered for.

• 1:09:14Lex Fridman?

  It's interesting that you spent time at AAAI on a panel with Garry Kasparov. What, I mean, it's almost, I'm just curious to learn the conversations you've had with Garry because he's also now, he's written a book about artificial intelligence. He's thinking about AI. He has kind of a view of it and he talks about AlphaGo a lot. What's your sense?

• 1:09:43Lex Fridman?

  Arguably, I'm not just being Russian, but I think Gary is the greatest chess player of all time, probably one of the greatest game players of all time. And you sort of at the center of creating a system that beats one of the greatest players of all time. So what's that conversation like? Is there anything, any interesting digs, any bets, any funny things, any profound things?

• 1:10:10Speaker 01

  So Gary Kasparov has an incredible respect for what we did with AlphaGo. And it's an amazing tribute coming from him of all people, that he really appreciates and respects what we've done. And I think he feels that the progress which has happened in computer chess, which later after AlphaGo, we built the AlphaZero system, which defeated the world's strongest chess programs. And to Gary Kasparov, that moment in computer chess was more profound than Deep Blue. And the reason he believes it mattered more was because

• 1:10:53Speaker 01

  it was done with learning and a system which was able to discover for itself new principles, new ideas, which were able to play the game in a way which he hadn't always known about, or anyone. And in fact, one of the things I discovered at this panel was that the current world champion, Magnus Carlsen, apparently recently commented on his improvement in performance, and he attributes it to AlphaZero. He's been studying the games of AlphaZero, and he's changed his style to play more like AlphaZero. And it's led to him actually increasing his rating to a new peak.

• 1:11:32Lex Fridman?

  Yeah, I guess to me, just like to Gary, the inspiring thing is that, and just like you said with reinforcement learning, reinforcement learning and deep learning, machine learning feels like what intelligence is.

• 1:11:46Speaker 01

  Yeah.

• 1:11:48Lex Fridman?

  You know, you could attribute it to sort of a bitter viewpoint from Gary's perspective, from us humans' perspective, saying that pure search that IBM Deep Blue was doing is not really intelligence, but somehow it didn't feel like it. And so that's the magical, I'm not sure what it is about learning that feels like intelligence, but it does.

• 1:12:11Speaker 01

  So I think we should not demean the achievements of what was done in previous eras of AI. I think that Deep Blue was an amazing achievement in itself and that heuristic search of the kind that was used by Deep Blue had some powerful ideas that were in there, but it also missed some things. So, so the fact that the, that the evaluation function, the way that the chess position was understood, was created by humans and not by the machine, is a limitation which means that there's a ceiling on how well it can do, but maybe more importantly, it means that the same idea cannot be applied in other domains where we don't have access

• 1:12:51Speaker 01

  to the kind of human grandmasters and that ability to kind of encode exactly their knowledge into an evaluation function. And the reality is that the story of AI is that, you know, most domains turn out to be of the second type, where, where knowledge is messy, it's hard to extract from experts or it isn't even available. And so, so we need to solve problems in a different way. And I think AlphaGo is a step towards solving things in a way which, which puts learning as a first-class citizen and says, systems need to understand for themselves how to understand the world, how to judge their, the value of, of, of any action that they

• 1:13:36Speaker 01

  might take within that world and any state they might find themselves in. And in order to do that,

• 1:13:44Speaker 01

  We make progress towards AI.

• 1:13:46Lex Fridman?

  Yeah, so one of the nice things about this, about taking a learning approach to the game of Go or game playing is that the things you learn, the things you figure out are actually going to be applicable to other problems that are real world problems. That's sort of that's ultimately, I mean, there's two really interesting things about AlphaGo. One is the science of it, just the science of learning, the science of intelligence. And then the other is, well, you're actually learning to figuring out how to build systems that would be potentially applicable in in other applications, medical autonomous vehicles, robotics, all I mean, it's just open the door to all kinds of applications.

• 1:14:27Lex Fridman?

  So the next incredible step, right, really the profound step is probably AlphaGo Zero. I mean, it's arguable, I kind of see them all as the same place, but really, and perhaps you were already thinking that AlphaGo Zero is the natural, it was always going to be the next step, but it's removing the reliance on human expert games for pre-training, as you mentioned. So how big of an intellectual leap was this that that self-play could achieve superhuman level performance in its own? And maybe could you also say, what is self-play? I kind of mentioned it a few times, but.

• 1:15:08Speaker 01

  So let me start with self-play. So the idea of self-play is something which is really about systems learning for themselves, but in the situation where there's more than one agent. And so if you're in a game and a game is played between two players, then self-play is really about understanding that game just by playing games against yourself rather than against any actual real opponent. And so it's a way to kind of discover strategies without having to actually need to go out and play against any particular human player, for example.

• 1:15:53Speaker 01

  The main idea of AlphaZero was really to try and step back from any of the knowledge that we put into the system and ask the question, is it possible to come up with a single elegant principle by which a system can learn for itself all of the knowledge which it requires to play a game such as Go? Importantly, by taking knowledge out, you not only make the system less brittle in the sense that perhaps the knowledge you were putting in was just getting in the way and maybe stopping the system learning for itself, but also you make it more general.

• 1:16:34Speaker 01

  The more knowledge you put in, the harder it is for a system to actually be placed, taken out of the system in which it's been designed and placed in some other system that maybe would need a completely different knowledge base to understand and perform well. And so the real goal here is to strip out all of the knowledge that we put in to the point that we can just plug it into something totally different. And that to me is really, you know, the promise of AI is that we can have systems such as that, which, you know, no matter what the goal is, no matter what goal we set to the system, we can come up with, we have an algorithm which can be placed into that world, into that environment, and can succeed in achieving that goal.

• 1:17:18Speaker 01

  And then that, that's, to me, is almost the essence of intelligence, if we can achieve that. And so AlphaZero is a step towards that. And it's a step that was taken in the context of two-player, perfect information games like Go and Chess. We also applied it to Japanese chess. So just to clarify, the first step was AlphaGo Zero? The first step was to try and take all of the knowledge out of AlphaGo in such a way that it could play in a fully self-discovered way, purely from self-play. And to me, the motivation for that was always that we could then plug it into other domains. But we saved that until later.

• 1:18:03Speaker 01

  Well, and... In fact, I mean... Just for fun, I could tell you exactly the moment where the idea for AlphaZero occurred to me, because I think there's maybe a lesson there for researchers who are kind of too deeply embedded in their research and, you know, working 24/7 to try and come up with the next idea, which is, it actually occurred to me on honeymoon, and I was like in my most fully relaxed state, really enjoying myself, and just bing, this, like, the algorithm for AlphaZero just appeared, like, and in its full form. And this was actually

• 1:18:47Speaker 01

  before we played against Lee Sedol, but we just didn't, I think we were so busy trying to make sure we could beat the world champion, that it was only later that we had the opportunity to step back and start examining that, that sort of deeper scientific question of whether this could really work.

• 1:19:09Lex Fridman?

  So nevertheless, so self-play is probably one of the most sort of profound ideas that represents, to me at least, artificial intelligence. But the fact that you could use that kind of mechanism to, again, beat world-class players, that's very surprising. So we kind of, to me, it feels like you have to train in a large number of expert games. So was it surprising to you? What was the intuition? Can you sort of think, not necessarily at that time, even now, what's your intuition? Why this thing works so well? Why it's able to learn from scratch?

• 1:19:48Speaker 01

  Well, let me first say why we tried it. So we tried it both because I feel that it was the deeper scientific question to be asking, to make progress towards AI, and also because in general, in my research... I don't like to do research on questions for which we already know the likely outcome. I don't see much value in running an experiment where you're 95% confident that you will succeed. And so we could have tried, you know, maybe to to take AlphaGo and do something which we we knew for sure it would succeed on. But much more interesting to me was to try try it on the things which we weren't sure about. And one of the big questions on our minds back then was, you know, could you really do this with self-play alone?

• 1:20:33Speaker 01

  How far could that go? Would it be as strong? And honestly, we weren't sure. Yeah, it was 50 50, I think, you know, we I really, if you'd asked me, I wasn't confident that it could reach the same level as these systems, but it felt like the right question to ask. And even if even if it had not achieved the same level, I felt that that was an important

• 1:20:58Speaker 01

  direction to be studying. And so

• 1:21:02Speaker 01

  then, lo and behold, it actually ended up outperforming the the previous version of of AlphaGo and indeed was able to beat it by 100 games to zero. So what's the intuition as to as to why? I think the intuition to me is clear that whenever you have errors in a in a system, as we did in AlphaGo, AlphaGo suffered from these delusions. Occasionally, it would misunderstand what was going on in a position and miscalculate it. How can how can you remove all of these these errors? Errors arise from many sources. For us, they were arising both from, you know, starting from the human data, but also from the from the nature of the search and the nature of the algorithm itself.

• 1:21:46Speaker 01

  But the only way to address them in any complex system is to give the system the ability to correct its own errors. It must be able to correct them. It must be able to learn for itself when it's doing something wrong and correct for it. And so it seemed to me that the way to correct delusions was indeed to have more iterations of reinforcement learning, that, you know, no matter where you start, you should be able to correct those errors until it gets to play that out and and understand, oh, well, I thought that I was going to win in this situation, but then I ended up losing. That suggests that I was misevaluating something. There's a hole in my knowledge and now now the system can correct for itself and and understand how to do better.

• 1:22:28Speaker 01

  Now, if you take that same idea and trace it back all the way to the beginning, it should be able to take you from no knowledge, from completely random starting point, all the way to the highest levels of knowledge that you can achieve in a domain. And the principle is the same, that if you bestow a system with the ability to correct its own errors, then it can take you from random to something slightly better than random, because it sees the stupid things that the random is doing, and it can correct them. And then it can take you from that slightly better system and understand, well, what's that doing wrong? And it takes you on to the next level and the next level. And this progress can go on indefinitely. And indeed, what would have happened if

• 1:23:12Speaker 01

  we'd carried on training AlphaZero for longer? We saw no sign of it slowing down its improvements, or at least it was certainly carrying on to improve. And presumably, if you had the computational resources, this could lead to better and better systems that discover more and more.

• 1:23:32Lex Fridman?

  So your intuition is fundamentally, there's not a ceiling to this process. One of the surprising things, just like you said, is the process of patching errors, it intuitively makes sense. Reinforcement learning should be part of that process. But what is surprising is, in the process of patching your own lack of knowledge, you don't open up other patches. You keep sort of, like, there's a monotonic decrease of your weaknesses.

• 1:24:05Speaker 01

  Well, let me back this up. I think science always should make falsifiable hypotheses.

• 1:24:10Lex Fridman?

  Yes.

• 1:24:10Speaker 01

  So let me back up this claim with a falsifiable hypothesis, which is that if someone was to, in the future, take AlphaZero as an algorithm and run it with greater computational resources that we had available today, then I would predict that they would be able to beat the previous system 100 games to zero. And that if they were then to do the same thing a couple of years later, that that would beat that previous system 100 games to zero, and that that process would continue indefinitely throughout at least my human lifetime.

• 1:24:44Lex Fridman?

  Presumably the game of Go would set the ceiling.

• 1:24:48Speaker 01

  I mean. The game of Go would set the ceiling, but the game of Go has 10 to the 170 states in it. So the ceiling is unreachable by any computational device that can be built out of the, you know, 10 to the 80 atoms in the universe. You asked a really good question, which is, you know, do you not open up other errors when you, when you correct your previous ones? And the answer is, is yes, you do. And so, so it's a remarkable fact about, about this class of, of two-player game and also true of single-agent games that essentially progress will always lead you to, if, if you have sufficient

• 1:25:30Speaker 01

  representational resource, like imagine you had, could represent every state in a big table of the game, then we, we know for sure that a progress of self-improvement will lead all the way in the single-agent case to the optimal possible behavior and in the two-player case to the minimax optimal behavior, that is the, the best way that I can play, knowing that you're playing perfectly against me.

• 1:25:55Speaker 01

  And so, so for those cases, we know that even if you do open up some new error, that in some sense you've made progress, you've, you've, you're progressing towards the, the best that can be done.

• 1:26:07Lex Fridman?

  So AlphaGo was initially trained on expert games with some self-play. AlphaGo Zero removed the need to be trained on expert games. And then another incredible step for me, because I just love chess, is to generalize that further to be in AlphaZero, to be able to play the game of Go, beating AlphaGo Zero and AlphaGo, and then also being able to play the, the game of chess and others. So what was that step like? What's the interesting aspects there that required to make that happen?

• 1:26:43Speaker 01

  I think the remarkable observation which we saw with AlphaZero was that actually without modifying the algorithm at all, it was able to play and crack some of AI's greatest previous challenges. In particular, we dropped it into the game of chess and unlike the previous systems like DeepBlue, which had been worked on for years and years, and we were able to beat the world's strongest computer chess program convincingly using a system that was fully discovered by its own, from scratch with its own principles. And in fact, one of the nice things that we found was that, in fact, we also achieved

• 1:27:27Speaker 01

  the same result in Japanese chess, a variant of chess where you get to capture pieces and then place them back down on your own side as an extra piece. So a much more complicated variant of chess. And we also beat the world's strongest programs and reached superhuman performance in that game too. And it was the very first time that we'd ever run the system on that particular game was the version that we published in the paper on AlphaZero. It just worked out of the box, literally. No touching it. We didn't have to do anything. And there it was, superhuman performance, no tweaking, no twiddling. And so I think there's something beautiful about that principle, that you can take an algorithm and without

• 1:28:10Speaker 01

  twiddling anything, it just works. Now, to go beyond AlphaZero, what's required? AlphaZero is just a step. And there's a long way to go beyond that to really crack the deep problems of AI. But one of the important steps is to acknowledge that the world is a really messy place. It's this rich, complex, beautiful, but messy environment that we live in. And no one gives us the rules. No one knows the rules of the world. At least, maybe we understand that it operates according to Newtonian or quantum mechanics at the micro level or according to relativity at the macro level. But that's not a model that's useful for us as people

• 1:28:55Speaker 01

  to operate in it. Somehow, the agent needs to understand the world for itself in a way where no one tells it the rules of the game, and yet it can still figure out what to do in that world. Deal with this stream of observations coming in, rich sensory input coming in, actions going out in a way that allows it to reason in the way that AlphaGo or AlphaZero can reason, in the way that these Go and chess playing programs can reason, but in a way that allows it to take actions in that messy world to achieve its goals. And so this led us to the most recent step in the story of AlphaGo, which was a system

• 1:29:35Speaker 01

  called MuZero, and MuZero is a system which learns for itself even when the rules are not given to it. It actually can be dropped into a system with messy perceptual inputs. We actually tried it in some Atari games, the canonical domains of Atari that have been used for reinforcement learning. And this system learned to build a model of these Atari games that was sufficiently rich and useful enough for it to be able to plan successfully. And in fact, that system not only went on to beat the state of the art in Atari, but the same system without modification was able to reach the same level of superhuman

• 1:30:19Speaker 01

  performance in Go, chess and Shogi that we'd seen in AlphaZero, showing that even without the rules, a system can learn for itself just by trial and error, just by playing this game of Go. And no one tells you what the rules are, but you just get to the end and someone says, you know, win or loss. You play this game of chess and someone says win or loss, or you play a game of Breakout in Atari and someone just tells you, you know, your score at the end. And the system for itself figures out essentially the rules of the system, the dynamics of the world, how the world works. And not in any explicit way, but just implicitly enough understanding for it to be able to plan in that system in order to achieve its goals.

• 1:31:02Lex Fridman?

  And that's the, you know, that's the fundamental process that you have to go through when you're facing any uncertain kind of environment that you would in the real world, is figuring out the sort of the rules, the basic rules of the game.

• 1:31:13Speaker 01

  That's right.

• 1:31:14Lex Fridman?

  So this, I mean, yeah, that allows it to be applicable to basically any domain that could be.

• 1:31:22Lex Fridman?

  Digitized in the way that it needs to in order to be consumable, sort of in order for the reinforcement learning framework to be able to sense the environment, to be able to act in the environment and so on.

• 1:31:32Speaker 01

  The full reinforcement learning problem needs to deal with with worlds that are unknown and and complex and and the agent needs to learn for itself how to deal with that. And so Mu zero was as a step, a further step in that direction.

• 1:31:46Lex Fridman?

  One of the things that inspired the general public and interesting conversations I have, like with my parents or something, my mom, that just loves what was done is kind of, at least the notion that there was some display of creativity, some new strategies, new behaviors that were created, that that again has echoes of intelligence. So is there something that stands up? Do you see it the same way that there's creativity and there's some behaviors, patterns that you saw that AlphaZero was able to display that are truly creative?

• 1:32:18Speaker 01

  So let me start by, I think, saying that I think we should ask what creativity really means. So to me, creativity means discovering something which wasn't known before, something unexpected, something out outside of our norms. And so in that sense, the process of reinforcement learning or the self-play approach that was used by AlphaZero is it's the essence of creativity. It's really saying at every stage, you're playing according to your current norms and you try something and if it works out, you say, hey, here's something great. I'm going to start using that.

• 1:33:01Speaker 01

  And then that process, it's like a micro discovery that happens millions and millions of times over the course of the algorithm's life, where it just discovers some new idea. Oh, this pattern, this pattern's working really well for me. I'm gonna, I'm gonna start using that. Oh, now, oh, here's this other thing I can do. I can start to to connect these stones together in this way, or I can start to, you know, sacrifice stones or give up on on on pieces or play shoulder hits on the fifth line or whatever it is. The system's discovering things like this for itself. Continually, repeatedly, all the time. And so it should come as no surprise to us then. When, if you leave these systems going, that they discover things that are not known to humans.

• 1:33:42Speaker 01

  That to the human norms are considered creative. And we've seen this several times. In fact, in AlphaGo Zero, we saw this beautiful timeline of discovery. Where what we saw was that there are these opening patterns that humans play called joseki. These are like the patterns that humans learn to play in the corners. And they've been developed and refined over, over literally thousands of years in the game of Go. And what we saw was in the course of the training, AlphaGo Zero, of the course of the 40 days that we trained this system. It starts to discover exactly these patterns that human players play.

• 1:34:23Speaker 01

  And over time, we found that all of the joseki that humans played were discovered by the system. Through this process of self-play and this sort of essential notion of creativity. But what was really interesting was that over time, it then starts to discard some of these in favor of its own joseki that humans didn't know about. And it starts to say, oh, well, you thought that the knight's move pinter joseki was a great idea.

• 1:34:52Speaker 01

  But here's something different you can do there, which makes some new variation that humans didn't know about. And actually, now the human Go players study the joseki that AlphaGo played. And they become the new norms that are used in today's top-level Go competitions.

• 1:35:08Lex Fridman?

  That never gets old, even just the first to me. Maybe just makes me feel good as a human being. That a self-play mechanism that knows nothing about us humans discovers patterns that we humans do. It's a affirmation that we're doing, we're doing okay as humans.

• 1:35:25Speaker 01

  Yeah.

• 1:35:26Lex Fridman?

  We've, in this domain and other domains, we figured out, it's like the Churchill quote about democracy. It's the, you know, it sucks, but it's the best one we've tried. So in general, taking a step outside of Go and you've like a million accomplishments that I have no time to talk about. With AlphaStar and so on, and the current work. But in general, this self-play mechanism that you've inspired the world with by beating the world champion Go player.

• 1:35:59Lex Fridman?

  Do you see that as, do you see it being applied in other domains? Do you have sort of dreams and hopes that it's applied in both the simulated environments and the constrained environments of games? Constrained, I mean, AlphaStar really demonstrates that you can remove a lot of the constraints, but nevertheless, it's in a digital simulated environment. Do you have a hope, a dream that it starts being applied in the robotics environment? And maybe even in domains that are safety critical and so on and have, you know, have a real impact in the real world, like autonomous vehicles, for example, which seems like a very far out dream at this point.

• 1:36:38Speaker 01

  So I absolutely do hope and imagine that we will, we will get to the point where ideas just like these are used in all kinds of different domains. In fact, one of the most satisfying things as a researcher is when you start to see other people use your algorithms in unexpected ways. So in the last couple of years, there have been, you know, a couple of Nature papers where different teams, unbeknownst to us, took AlphaZero and applied exactly those same algorithms and ideas to real world problems of huge meaning to society. So one of them was the problem of chemical synthesis, and they were able to beat the state of the art in finding pathways

• 1:37:21Speaker 01

  of how to actually synthesize chemicals, retro chemical synthesis. And the second paper, actually just came out a couple of weeks ago in Nature, showed that in quantum computation, you know, one of the big questions is how to, how to understand the nature of the function in quantum computation. And a system based on AlphaZero beat the state of the art by quite some distance there again. So, so these are just examples, and I think, you know, the lesson which we've seen elsewhere in machine learning, time and time again, is that if you make something general, it will be used in all kinds of ways. You know, you provide a really powerful tool to society,

• 1:38:04Speaker 01

  and, and those tools can be used in amazing ways. And so I think we're just at the beginning, and, and for sure, I hope that we see all kinds of outcomes.

• 1:38:15Lex Fridman?

  So the, and the other side of the question of reinforcement learning framework is, you know, you usually want to specify a reward function and an objective function. What do you think about sort of ideas of intrinsic rewards if, and when we're not really sure about, you know, of, if we take, you know, human beings as existence proof, that we don't seem to be operating according to a single reward, do you think that there's interesting ideas for when you don't know how to truly specify the reward, you know, that there's some flexibility for discovering it intrinsically or so on in the context of reinforcement learning?

• 1:38:59Speaker 01

  So I think, you know, when we think about intelligence, it's really important to be clear about the problem of intelligence. And I think it's clearest to understand that problem in terms of some ultimate goal that we want the system to try and solve for. And after all, if we don't understand the ultimate purpose of the system, do we really even have a clearly defined problem that we're solving at all? Now, within that, as with your example for humans, the system may choose to create its own motivations and subgoals that help the system to achieve its ultimate goal. And that may indeed be a hugely important mechanism to achieve those ultimate goals. But there is still some ultimate goal I think

• 1:39:42Speaker 01

  the system needs to be measurable and evaluated against. And even for humans, I mean, humans, we're incredibly flexible. We feel that we can, you know, any goal that we're given, we feel we can, we can master to some degree. But if we think of those goals, really, you know, like the goal of being able to pick up an object, or the goal of being able to communicate, or the influence people to do things in a particular way, or whatever those goals are, really, they are, they're subgoals, really, that we set ourselves. You know, we choose to pick up the object, we choose to communicate, we choose to, to influence someone else. And we choose those because we think it will lead us to something in

• 1:40:26Speaker 01

  our, in later on. We think that that's helpful to us to achieve some ultimate goal. Now, I don't want to speculate whether or not humans as a system necessarily have a singular overall goal of survival or whatever it is. But I think the principle for understanding and implementing intelligence is has to be that if we're trying to understand intelligence or implement our own, there has to be a well-defined problem. Otherwise, If it's not, I think it's like an admission of defeat. That for there to be hope for for understanding or implementing intelligence, we have to know what we're doing. We have to know what we're asking the system to do. Otherwise, if you if you don't have a clearly defined purpose, you're not going to get a clearly defined answer.

• 1:41:08Lex Fridman?

  The ridiculous big question that has to naturally follow is I have to pin you down on this on this thing that nevertheless, one of the big silly or big real questions before humans is the meaning of life, is us trying to figure out our own reward function.

• 1:41:27Speaker 01

  Yeah.

• 1:41:28Lex Fridman?

  And you just kind of mentioned that if you want to build intelligent systems and you know what you're doing, you should be at least cognizant to some degree of what the reward function is. So the natural question is, what do you think is the reward function of human life? The meaning of life for us humans? The meaning of our existence?

• 1:41:49Speaker 01

  I think, you know, I'd be speculating beyond my own expertise, but but just for fun, let me do that.

• 1:41:55Lex Fridman?

  Yes, please.

• 1:41:56Speaker 01

  And say, I think that there are many levels at which you can understand a system and and you can understand something as as optimizing for for a goal at many levels. And so so you can understand the the, you know, let's start with the universe. Like, does the universe have a purpose? Well, it feels like it's just at one level, just following certain mechanical laws of physics and that that's led to the development of the universe. But at another level, you can view it as actually there's the second law of thermodynamics that says that this is increasing in entropy over time forever. And now there's a view that's been developed by certain people at MIT that this you can think of this as as almost like a goal of the universe, that the purpose of the universe is to maximize entropy.

• 1:42:41Speaker 01

  So there are multiple levels at which you can understand a system. The next level down, you might say, well, if the goal is to is to maximize entropy, well, how do you? How does, how can that be done by a particular system? And maybe evolution is something that the universe discovered in order, in order to kind of dissipate energy as efficiently as possible. And by the way, I'm borrowing from Max Tegmark for some of these metaphors. Yes. The physicist. But if you can think of evolution as a mechanism for dispersing energy, then, then evolution, you might say is, then becomes a goal, which is if, if evolution disperses

• 1:43:23Speaker 01

  energy by reproducing as efficiently as possible, what's evolution then? Well, it's now got its own goal within that, which is to actually reproduce as effectively as possible. And now, how does reproduction, how is that made as effective as possible?

• 1:43:42Speaker 01

  Well, you need entities within that that can survive and reproduce as effectively as possible. And so it's natural that in order to achieve that high level goal, those individual organisms discover brains, intelligences, which enable them to support the, the goals of, of evolution. And those brains, what do they do? Well, perhaps the early brains, maybe they were controlling things at some direct level. You know, maybe they were the equivalent of pre-programmed systems, which were directly controlling what was going on and setting certain, you know, things in order to achieve these particular, particular goals. But that led to another level of, of discovery, which was learning systems, you know, parts

• 1:44:24Speaker 01

  of the brain which are able to, to learn for themselves and learn how to, to program themselves to achieve any goal. And presumably there are parts of the, of the brain where goals are set to, to parts of that, that, that system and provides this very flexible notion of intelligence that, that we as humans presumably have, which is the ability to kind of, why the reason we feel that we can, we can, we can achieve any goal. So, so it's a very long-winded answer to say that, you know, I think there are many perspectives and many levels at which intelligence can be understood. And, and each of those levels, you can take multiple perspectives, like, you know, you can view the system as, as something which is optimizing for a goal, which is understanding.

• 1:45:03Speaker 01

  At a level by which we can maybe implement it and understand it as AI researchers or computer scientists. Or you can understand it at the level of the mechanistic thing which is going on, that there are these atoms bouncing around in the brain and they lead to the outcome of that system. It's not in contradiction with the fact that it's also a decision making system that's optimizing for some goal and purpose.

• 1:45:27Lex Fridman?

  I've never heard the description of the meaning of life structured so beautifully in layers. But you did miss one layer, which is the next step, which you're responsible for, which is creating the artificial intelligence.

• 1:45:43Speaker 01

  Indeed.

• 1:45:43Lex Fridman?

  Layer on top of that.

• 1:45:45Speaker 01

  Indeed.

• 1:45:46Lex Fridman?

  I can't wait to see, well, I may not be around, but I can't wait to see what the next layer beyond that will be.

• 1:45:53Speaker 01

  Well, let's just take that that argument and pursue it to its natural conclusion. So the next level indeed is for how can our how can our learning brain achieve its goals most effectively? Well, maybe it does so by us as learning beings, building a system which is able to solve for those goals more effectively than we can. And so when we build a system to play the game of Go, you know, when I said that I wanted to build a system that can play Go better than I can. I've enabled myself to achieve that goal of playing Go better than I could by directly playing it and learning it myself.

• 1:46:32Speaker 01

  And so now a new layer has been created, which is systems which are able to achieve goals for themselves. And ultimately, there may be layers beyond that where they set subgoals to parts of their own system in order to achieve those and so forth. So.

• 1:46:50Lex Fridman?

  Incredible.

• 1:46:50Speaker 01

  So the story of intelligence, I think, I think is a multi-layered one and a multi-perspective one.

• 1:46:56Lex Fridman?

  We live in an incredible universe. David, thank you so much, first of all, for dreaming of using learning to solve Go and building intelligent systems. And for actually making it happen and for inspiring millions of people in the process. It's truly an honor. Thank you so much for talking today.

• 1:47:15Speaker 01

  Okay, thank you.

• 1:47:16Lex Fridman?

  Thanks for listening to this conversation with David Silver, and thank you to our sponsors, MasterClass and Cash App. Please consider supporting the podcast by signing up to [email protected]/lex and downloading Cash App and using code LEXPODCAST. If you enjoy this podcast, subscribe on YouTube, review it with five stars on Apple Podcasts, support it on Patreon, or simply connect with me on Twitter at @LexFriedman. And now, let me leave you with some words from David Silver. My personal belief is that we've seen something of a turning point, where we're starting to understand that many abilities, like intuition and creativity, that we've previously thought were in the domain only of the human mind,

• 1:47:58Lex Fridman?

  are actually accessible to machine intelligence as well. And I think that's a really exciting moment in history. Thank you for listening and hope to see you next time.