YouSaid · the spoken record

Tuomas Sandholm

lines on the record: 72
first: 2018-12-28
most recent: 2018-12-28
sittings or episodes: 1
sources: podcast

Every line below is reproduced as it was said and linked to the record it came from. Nothing here is summarised or generated. Directory · Search · Corrections

“So, which is kind of interesting that algorithm configuration has been going on now for at least 17 years seriously, and there has not been any generalization theory before.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Let's say business strategy and not just modeling like a particular interaction, but thinking about the business from here to eternity Or let's say military strategy. So it's not like war is going to go away. How do you think about military strategy that's going to go forever? How do you even model that? How do you know whether a move was good that somebody made and so on? So that's kind of one direction. I'm also very interested in learning much more scalable techniques for integer programming. So we had a nice email paper this summer on that. The first automated algorithm configuration paper that has theoretical generalization guarantees. So if I see this many training examples and I turn my algorithm in this way, it's going to have good performance on the real distribution, which I have not seen.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Yeah, yeah, lots of different things in the game solving. So solving even bigger games, games where you have more hidden action of the player actions as well. Poker is a game where really the chance actions are hidden, or some of them are hidden, but the player actions are public. Multiplayer games of various sorts, collusion, opponent exploitation, and even longer games. So games that basically go forever, but they're not repeated. Seek extensive on games that go forever, what would that even look like? How do you represent that? How do you solve that?”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Yeah, I think that's always a challenge. That kind of slowness and inertia, really, of let's do things the way we've always done it. You just have to find the internal champions at the customer who understand that, hey, things can't be the same way in the future. Otherwise, bad things are going to happen. And it's in autonomous vehicles. It's actually very interesting that the car makers are doing that, and they're very traditional. But at the same time, you have tech companies who have nothing to do with cars or transportation, like Google and Baidu really pushing on autonomous cars. I find that fascinating.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“The number one thing for me right now is coming up with these scalable techniques for game solving and applying them into the real world. I'm still very interested in market design as well, and we're doing that in the optimized markets. But I'm most interested if, number one right now is strategic machine strategy robot, getting that technology out there and seeing as you're in the trenches doing applications what needs to be actually filled, what technology gaps still need to be filled. So it's so hard to just put your feet on the table and imagine what needs to be done. But when you're actually doing real applications, the applications tell you what needs to be done. And I really enjoy that interaction.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“I am still extremely worried. So you probably know the simple game theory of MAD. So this was a mutually assured destruction. And it doesn't require any computation with small matrices. You can actually convince yourself that the game is such that nobody wants to initiate. Yeah, that's very coarse grained analysis and it really works in a situation of where you have two superpowers or small number of superpowers. Now things are very different. You have a smaller Nook so the threshold of initiating is smaller and you have smaller countries and non-nation actors who may get nooks and so on. So I think it's riskier now than it was maybe ever before.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“political will and in the US you know the market at least the Chicago market was just shut down and so on so then it doesn't really help how great your market design was”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“So, I do think that about that a lot. I think the biggest two threats that we're facing as mankind, one is climate change, and the other is nuclear war. So those are my main worries that I worry about. And I've tried to do something about climate, thought about trying to do something for climate change twice, actually. for two of my startups. I've actually commissioned studies of what we could do on those things and we didn't really find a sweet spot, but I'm still keeping an eye out on that. If there's something where we could actually provide a market solution or optimization solution or some other technology solution to problems, right now, like, for example, pollution credit markets was what we were looking at then. And it was much more the lack of political will by those markets were not so successful rather than bad market design. So I could go in and make a better market design, but that wouldn't really move the needle on the world very much if there's”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Wrong objective, the AI will optimize that to the hilt and it's going to hurt more than some human who's kind of trying to Solid and a half baked way with some human insight too, but I just haven't seen that materialize in practice.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Yeah, I think this value misalignment is a fairly theoretical worry. And I haven't really seen it anymore because I do a lot of real applications. I don't see it anywhere. The closest I've seen it was the following type of mental exercise, really, where I had this argument in the late 80s when we were building these transportation optimization systems. And somebody had heard that it's a good idea to have high utilization of assets. So they told me that, hey, why don't you put that as objective? And we didn't even put it as an objective because I just showed him, you know, if you had that as your objective, the solution would be to load your trucks full and drive in circles. Nothing would ever get delivered. You'd have 100% utilization. So yeah, I know this phenomenon. I've known this for over 30 years, but I've never seen it actually be a problem in reality. And yes, if you have...”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Let me actually come back on 30 as one thing. I think AI is also going to make the world much safer. So that's another aspect that often gets overlooked.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“So that's over $6 billion of efficiency improvement in the world. And this is not like shifting value from somebody to somebody else, just efficiency improvement, like in trucking, less empty driving. So there's less. Waste less carbon footprint and so on.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“I am much more optimistic about the positive impacts. So just in my own work, what we've done so far, we run the nationwide kidney exchange. Hundreds of people are walking around alive today, who would it be? And it's increased employment. You have a lot of people now running kidney exchanges and at transplant centers interacting with the kidney exchange. You have extra surgeons, nurses, anesthesiologists, hospitals, all of that. So employment is increasing from that and the world is becoming a better place. Another example is combinatorial sourcing auctions. We did 800 large-scale combinatorial sourcing auctions from 2001 to 2010 in a previous startup of mine called Combine Net. And we increased the supply chain efficiency on that $60 billion of spend by”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“So that's a good question. I would say a little bit yes and no. And what I mean by that is that these game theoretic strategies, let's say Nash equilibrium, it has provable properties. So it's unlike, let's say, deep learning where... Now, that doesn't necessarily mean that the strategies are human understandable. That's a whole other problem. So I think that deep learning and computational game theory are in the same boat in that sense, that both are difficult to understand. But at least the game theoretic techniques, they have these guarantees of solar quality.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Now, I think the next challenge problem, I know you're not asking about it this way, you're asking about technology breakthrough. But I think the big, big breakthrough is to be able to show it, hey, maybe most of, let's say, military planning or most of business strategy will actually be done strategically using computational game theory. That's what I would like to see as a next five or ten-year goal.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Yeah, so I think overall, it's We're in a very different situation in game theory than we are in, let's say, machine learning So in machine learning, it's a fairly mature technology and it's very broadly applied and proven success in the real world. In game solving, there are almost no applications yet. We have just become superhuman, which machine learning you could argue happened in the 90s, if not earlier, and at least on supervised learning at certain complex supervised learning applications.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Yeah, so that's kind of how I am. So I am probably not going to work as hard on these recreational benchmarks. I'm doing two startups on game solving technology, strategic machine and strategy robot. And we're really interested in pushing this stuff into practice.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Or you could have diplomacy or Hanabi, you know, things like that. These are like recreational games, but none of them are really acknowledged as kind of the main next challenge problem, like chess or go or heads up no limit, Texas Holden was. So I don't really know in the game solving space what is or what will be the next benchmark. I kind of hope that there will be a next benchmark because really the different groups working on the same problem really drove this application independent techniques forward very quickly over 10 years.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“So that's a great question, and I don't really know the answer in terms of game solving, heads up no limit text really was the one remaining widely agreed upon benchmark. So that was the big milestone. Now, are there other things? Yeah, certainly there are, but there's not one that the community has kind of focused on. So what could be other things? There are groups working on StarCraft. There are groups working on Dota 2. These are video games.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Not so much a lot of players, but many items for sale. And these mechanisms are such that even with just two items or one item, bidding truthfully wouldn't be the best strategy.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“They're such that bidding truthfully is not the best strategy. Usually mechanism design we try to make things easy for the participants. So telling the truth is the best strategy. But even in those very high-stakes auctions where you have tens of billions of dollars worth of spectrum being auctioned. Truth telling is not the best strategy And by the way, nobody knows even a single optimal bidding strategy for those auctions.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“So, mechanism design itself has had Fairly limited success so far. There are certain cases, but most of the real world situations are actually not sound from a mechanism design perspective. Even in those cases where they've been designed by very knowledgeable mechanism design people, the people are typically just taking some insights from the theory and applying those insights into the real world rather than applying the mechanisms directly. So one famous example of is the FCC Spectrum auctions. So I've also had a small role in that and very good economists have been working, excellent economists have been working on that who know game theory. Yet the rules that are designed in practice there”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Depending how you design the game. And of course, it doesn't in any way contradict the impossibility result. The impossibility result is still there, but it just finds spots within this impossible class where in those spots you don't have the impossibility.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Good thing about automated mechanism design is that we're not really designing for a class, we're designing for specific settings at a time. So even if there's an impossibility result for the whole class, it just doesn't mean that all of the cases in the class are impossible. It just means that some of the cases are impossible. So we can actually carve these islands of possibility within these known impossible classes. And we've actually done that. So one of the famous results in mechanism design is some Myers and Sadethweight theorem by Roger Myerson and Mark Satetwaite from 1983. It's an impossibility of efficient trade under imperfect information. We show that you can in many settings avoid that and get efficient trade anyway.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Impossible. So these are not statements about human ingenuity, who might come up with something smart. These are proofs that if you want to accomplish properties X in class C, that is not doable with any mechanism.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“But it's not a panacea. There are impossibility results in mechanism design, saying that there's no mechanism that accomplishes objective X in class C. So there's no way using any mechanism design tools, manual or automated, to do certain things in the mechanism design.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“So in general, I believe it's very important to do things in the real world and at scale. And that's really where the The pudding, if you will, proof is in the pudding. That's where it is. In this particular case, it was kind of a competition between different groups for many years as to who can be the first one to be the top humans at heads up Nolimitx as Holden. So it became kind of a competition who can get there.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“I don't know about Brave, but it takes a lot of work. It takes a lot of work and a lot of time to organize, do make something big and to organize an event and stuff like that”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“That's what I think. And I think the same thing happens in poker. And so I didn't think of myself as somebody who's going to kill the game. And I don't think I did. I've really learned to love this game. I wasn't a poker player before, but learned so many nuances about from these AIs. And they've really changed how the game is played, by the way. So they have this very Martian ways of playing poker, and the top humans are now incorporating those types of strategies into their own play. So if anything to me, our work has made poker a richer, more interesting game for humans to play, not something that is going to steer humans away from it entirely.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Yeah, a lot of people felt that it was a real threat to the whole game, the whole existence of the game. If AI becomes better than humans, people would be scared to play poker because there are these superhuman AIs running around taking their money and all of that. So I just, it's just really aggressive. The comments were super aggressive. I got everything just short of death threats.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“And I really mean in terms of convergence rates better, like first order methods, better convergence rates like the CFR-based algorithms, yet the CFR-based algorithms are the fastest in practice. So it really tells me that you have to test this in reality. The theory isn't tight enough, if you will, to tell you which algorithms are better than the others. And you have to look at these things in the large because any sort of projections you do from the small can at least in this domain be very misleading. So that's kind of from a kind of a science and engineering perspective. From a personal perspective, it's been just a wild experience in that with the first poker competition, the first brains versus AI, man-machine. Poker competition that we organized. There had been, by the way, for other poker games. There had been previous competitions, but this was for heads up no limit. This was the first. And I probably became the one.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Yeah, so that's a good question. So there's so much to say about it. I do like this type of performance-oriented research, although in my group we go all the way from like idea to theory, to experiments to big system fielding to commercialization. So we span that spectrum. But I think that in a lot of situations in AI, you really have to build the big systems and evaluate them at scale before you know what works and doesn't. And we've seen that in the computational game theory community, that there are a lot of techniques that look good in the small. But then the seas to look good in the lodge. And we've also seen that there are a lot of techniques that look superior in theory.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“To negotiate all of these situations, or many of them, in advance. And of course, it might be that, hey, maybe you're not going to always let me go first. Maybe you said, okay, well, in these situations, I'll let you go first, but in exchange, you're going to give me to Amazon. You're going to let me go first in these situations. So it's this huge combinatorial negotiation.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Okay, so I haven't really thought about j-walking, but one thing that I think could be a good application in autonomous vehicles is the following. So let's say that you have fleets of autonomous cars operating by different companies. So maybe here's the way more fleet and here's the Uber fleet. If you think about the rules of the road, they define certain legal rules, but that still leaves a huge strategy space open. Like as a simple example, when cars merge, you know, how humans merge, you know, they slow down and look at each other and try to merge. Wouldn't it be better if these situations would already be pre-negotiated so we can actually merge at full speed and we know that this is a situation, this is how we do it, and it's all going to be faster. But there are way too many situations to negotiate manually. So you could use automated negotiation. This is the idea at least. You could use automated negotiation.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“I am not sure the goal really is modeling humans. Like, for example, if I'm playing a zero-sum game. I don't really care that the opponent is actually following my model of rational behavior. Because if they're not, that's even better for me.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“That's another company Optimized Markets. But that's much more about combinatorial market and optimization-based technology. That's not using these game theoretic reasoning technologies”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Yeah, I actually have two startup companies doing exactly that. One is called Strategic Machine, and that's for kind of business applications, gaming, sports, all sorts of things like that. Any applications of this to business and to sports and to gaming. Various types of things in finance, electricity markets, and so on. And the other is called strategy robot, where we are taking this to military, security, cybersecurity, and intelligence applications.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“So, like bridge. So think about bridge. It's like when you and I are on a team, our payoffs are the same. Problem is that we can't talk. So when I get my cards, I can't whisper to you what my cards are, that would not be allowed. So we have to somehow coordinate our strategies ahead of time and only ahead of time. And then there are certain signals we can talk about, but they have to be such that the other team also understands them. So that's an example where the coordination is already built into the rules of the game. But in many other situations like auctions or negotiations or diplomatic relationships, poker, it's not really built in, but it still can be very helpful for the colluders.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Yeah, so I've done a lot of work, coalitional games, and we actually have a paper here with my other student, Gabriella Farina, and some other collaborators at NIPS on that. Actually, just came back from the poster session where we presented this. So when you have a collusion, it's a different problem. And it typically gets even harder then. Even the game representations, some of the game representations don't really allow. computation so we actually introduced a new game representation for for that”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Okay, so it is true that all finite games have a Nash equilibrium. So this is what John Nash actually proved. So, they do have an ash equilibrium. That's not the problem. The problem is that there can be many. And then there's a question of which equilibrium to select. And if you select your strategy from a different equilibrium and I select mine, Then what does that mean? And in this non-zero-sum games, we may lose some. Benefit by being just simply stupid, we could actually both be better off if we did something else. And in three player, you get other problems also like collusion. Radically better by colluding. So there are lots of issues that come up there.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Even without cooperation. So there's a big gap from two player zeros to two player general sum or even to three player zero sum. That's a big gap. At least in theory.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“But they're conceptually easier and computationally easier in that conceptually you don't have to worry about which equilibrium is the other guy going to play when there are multiple because any equilibrium strategy is the best response to any other equilibrium strategy. So I can play a different equilibrium from you, and we'll still get the right values of the game. Falls apart even with two players when you have general sum games.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“A matrix, yeah, and that the matrix is called the matrix form or bimatrix for more normal form game. And here you have the tree form, so you can actually do certain types of reasoning there that you lose the information when you go to normal phone. There's a certain form of equivalence, like if you go from three form and you say every possible contingency plan. Is a strategy, then I can actually go back to the normal form, but I lose some information, lack of sequentiality. Then the multiplayer versus two-player distinction is an important one. So two-player games. Zero sum are conceptually easier computationally easier the still huge like this one this one”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Then if you move up from just repeated simple repeated matrix games, not all the way to extensive form games, but in between the stochastic games where you think about it like these little matrix games and when you take an action and your takes an action, they determine not which next state I'm going to next game I'm going to, but the distribution over next games where I might be going to. So that's the stochastic game. But it's like Matrix games repeated. Stochastic games extensive form games. That is from less to more general. And poker is an example of the last one. So it's really in the most general setting. Extension form games. And that's kind of what the AI community has been working on and being benchmarked on with this heads up Nolimitex Hold.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“What are the key differences? So let me start from the basics so A repeated game is a game where the same exact game is played over and over. In these extensive form games, where think about three form, maybe with these information sets to represent incomplete information, you can have kind of repetitive interactions and even repeated games are a special case of that, by the way. But the game doesn't have to be exactly the same. So like in sourcing auctions, yes, we're going to see the same supply-based year to year, but what I'm buying is a little different every time, and the supply base is a little different every time and so on. So it's not really repeated. So to find a purely repeated game is actually very rare in the world. So they're really a very coarse model of what's going on.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“That you can think about it this way a game theoretic strategy is unbeatable, but it doesn't maximally beat the other opponents. So the winnings per hand might be better with a different strategy. And the hybrid is that you start from a game theoretic approach, and then as you gain data from about the opponent, in certain parts of the game tree, then in those parts of the game tree, you start to tweak your strategy more and more towards exploitation while still staying fairly close to the game theory strategy so as to not open yourself up to exploitation too much.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Yeah, definitely. We've done some work on that and I really like the work that hybridizes the two. So you figure out what were the Rational opponent do. And by the way, that's safe in these zero-sum games, two player zero-sum games, because if the opponent does something irrational, yes, it might throw off my beliefs, but the amount that the player can gain by throwing off my belief is always less than they lose by playing poorly. So it's safe. But still, if somebody's weak as a player, you might want to play differently to exploit them more.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“No, the game theory isn't really player specific. So that's also why we don't need any data. We don't need any history how particular humans played in the past or how any AI or human had played before. It's all about rationality. So we just think the AI just thinks about what would a rational opponent do and what would I do if I were I am rational and that's the idea of game theory. So it's really a data free, opponent free.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“I'll do a simple example. So, you know, the game paper scissors, so we can draw it as player one moves first, and then player two moves. But of course, it's important that player two doesn't know what player one moved. Otherwise, player two would win every time. So, we can draw that as an information set where player one makes one of three moves first, and then there's an information set for player two. So player two doesn't know which of those nodes world is in. Once we know the strategy for player one, Nash equilibrium will say that you play one third rock, one third paper, one third scissors. From that, I can derive my beliefs on the information set that they're one-third, one-third, one-third.”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source
“Well, that's where this beauty of game theory comes. So Nash equilibrium, which John Nash introduced in 1950, introduces what rational play is when you have more than one player. And these are pairs of strategies where strategies are contingency plans, one for each player, so that neither player wants to deviate to a different strategy, given that the other doesn't deviate, but as a side effect, you get the beliefs from base role. So Nash equilibrium really isn't just deriving in this imperfect information games. Nash equilibrium doesn't just define strategies. It also defines beliefs for both of us and defines beliefs for each state. So each state called information sets. At each information set in the game, there's a set of different states. We might be in, but I don't know which one we're in”
2018-12-28 · Lex Fridman Podcast · Tuomas Sandholm: Poker and Game Theory · IDENTIFIED FROM THE TRANSCRIPT · source