Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

CMU's Libratus builds substantial lead in Brains vs. AI competition

cmu.edu

Post reply on HN
Print view View original

31–40 of 123 posts

Re: CMU's Libratus builds substantial lead in Brains vs. AI competition

#31

Post by nopinsight »

I believe we are witnessing the Cambrian explosion of intelligences. The techniques behind Libratus (abstraction algorithm and game theory [1]) appear to be qualitatively distinct from those behind AlphaGo (deep reinforcement learning and MCMC) and DeepBlue (search and heuristics).

An ecology of Artificial Intelligences, unbounded by our evolutionary history and neural architecture, could evolve to suit each particular task more effectively than our brains can.

Promises and perils abound.

[1] http://www.cs.cmu.edu/~sandholm/ section "Algorithms and complexity of solving games"

Re: CMU's Libratus builds substantial lead in Brains vs. AI competition

#32

Post by erik998 »

skoutus wrote: post #29

I wonder if the players tried colluding: coordinate their bets to fake a weakness so the AI start to adopt a poor strategy, then up the bets and stop the feign. I don't see how the AI can protect itself against that.

Isn't coordination between players what really happens in most poker games...

Re: CMU's Libratus builds substantial lead in Brains vs. AI competition

#34

Post by sagivo »

osti wrote: post #24
sagivo wrote: post #21

I would love to know where they got their training data or if there's one publicly available.

This is not neural net. It uses an algorithm called counterfactual regret minimization to compute the Nash equilibrium of the game, no data required.

what about - https://arxiv.org/abs/1701.01724

Re: CMU's Libratus builds substantial lead in Brains vs. AI competition

#35

Post by skoutus »

erik998 wrote: post #32
skoutus wrote: post #29

I wonder if the players tried colluding: coordinate their bets to fake a weakness so the AI start to adopt a poor strategy, then up the bets and stop the feign. I don't see how the AI can protect itself against that.

Isn't coordination between players what really happens in most poker games...

This is a multi-day competition. Players may coordinate in each round, but I mean to coordinate in each day. If you only coordinate over a short horizon then the complexity of your deception is lower. So for example, all players adopt a common feigned weakness on a given day, and let the computer believe those behaviors is part of a pattern to exploit. Then on the second day, up the bets and stop the feign.

This happens in algorithmic trading, where traders would make a large number of low-valued, bad bets to mislead the algo. Then bet big and go the other direction.

Re: CMU's Libratus builds substantial lead in Brains vs. AI competition

#36

Post by skoutus »

erik998 wrote: post #32
skoutus wrote: post #29

I wonder if the players tried colluding: coordinate their bets to fake a weakness so the AI start to adopt a poor strategy, then up the bets and stop the feign. I don't see how the AI can protect itself against that.

Isn't coordination between players what really happens in most poker games...

I just looked it up and collusion is illegal in poker, though players could do it.

Re: CMU's Libratus builds substantial lead in Brains vs. AI competition

#37

Post by hunl »

skoutus wrote: post #29

I wonder if the players tried colluding: coordinate their bets to fake a weakness so the AI start to adopt a poor strategy, then up the bets and stop the feign. I don't see how the AI can protect itself against that.

The aim of the AI isn't to adopt to poor strategies, rather to play an approximate optimal strategy itself. It's aiming to be unexploitable, the further the other players deviate from optimal, the more it wins. It's EV (expected value) comes from the other players not playing optimally, it doesn't care about exploiting individual weaknesses.

Re: CMU's Libratus builds substantial lead in Brains vs. AI competition

#38

Post by 36bydesignBL »

LAMike wrote: post #19

Wonder how long it will be until there is a crowdsourced AI winning the World Series of Poker?

Really long. It will still have to get lucky, or at least avoid being unlucky. This is why amateurs routinely crash the final table instead of just pros.

Re: CMU's Libratus builds substantial lead in Brains vs. AI competition

#39

Post by skoutus »

hunl wrote: post #37
skoutus wrote: post #29

I wonder if the players tried colluding: coordinate their bets to fake a weakness so the AI start to adopt a poor strategy, then up the bets and stop the feign. I don't see how the AI can protect itself against that.

The aim of the AI isn't to adopt to poor strategies, rather to play an approximate optimal strategy itself. It's aiming to be unexploitable, the further the other players deviate from optimal, the more it wins. It's EV (expected value) comes from the other players not playing optimally, it doesn't care about exploiting individual weaknesses.

AI's aim is not to adopt poor strategies. It seeks to optimize, but the point is you can lead it to believe a point is global optimum, when in fact it is only local optimum.

As to your point about the EV, this is why collusion can work. By colluding over a long enough horizon, the AI can believe that the average expected value to be something that it is not. If only one individual feign a weakness and the rest do not, then the strategy doesn't work.

Re: CMU's Libratus builds substantial lead in Brains vs. AI competition

#40

Post by hunl »

skoutus wrote: post #39
hunl wrote: post #37

Earlier quoted context omitted.

The aim of the AI isn't to adopt to poor strategies, rather to play an approximate optimal strategy itself. It's aiming to be unexploitable, the further the other players deviate from optimal, the more it wins. It's EV (expected value) comes from the other players not playing optimally, it doesn't care about exploiting individual weaknesses.

AI's aim is not to adopt poor strategies. It seeks to optimize, but the point is you can lead it to believe a point is global optimum, when in fact it is only local optimum. As to your point about the EV, this is why collusion can work. By colluding over a long enough horizon, the AI can believe that the average expected value to be something that it is not. If only one individual feign a weakness and the rest do not…

You can't lead it to believe a point is an optimum, it's just responded to a bet size/check in isolation given the information it has. If you 'feign weakness' in a given spot it will just respond as optimally as possible to the bet size.

For example attempting to feign weakness by betting small in a spot where your entire range should bet large is not tricking the AI, it's just passing up on EV for the players, good players are not going to play poorly in hope of tricking the bot for future mythical EV gain.

Also, there is no 'colluding' in heads up poker.

Post reply on HN