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

Grandmaster-level chess without search

github.com

Post reply on HN
Print view View original

11–20 of 171 posts

Re: Grandmaster-level chess without search

#11

Post by janalsncm »

alfalfasprout wrote: post #5
joelthelion wrote: post #3

I wonder if you could creatively combine this model with search algorithms to advance the state of the art in computer chess? I wouldn't be surprised to see such a bot pop up on tcec in a couple years.

The thing is classical chess (unlike eg; go) is essentially "solved" when run on computers capable of extreme depth. Modern chess engines play essentially flawlessly.

Chess is not “solved”. Solved doesn’t mean computers can beat humans, it means for any chess board position we can tell whether white wins, black wins, or the game is drawn with perfect play. We would know if the starting position was drawn, for example.

No computers now or in the foreseeable future will be capable of solving chess. It has an average branching factor over 30 and games can be over 100 moves.

Re: Grandmaster-level chess without search

#12

Post by janalsncm »

joelthelion wrote: post #3

I wonder if you could creatively combine this model with search algorithms to advance the state of the art in computer chess? I wouldn't be surprised to see such a bot pop up on tcec in a couple years.

The advantage of this flavor of engine is that it might make parallel position evaluation extremely efficient. Calculate 1024 leaf positions and batch them to the model, take the top 10% and explore their sub-trees either via further GPU batching or minimax eval.

NNUE already tries to distill a subtree eval into a neural net, but it’s optimized for CPU rather than GPU.

Re: Grandmaster-level chess without search

#13

Post by primitivesuave »

alfalfasprout wrote: post #5
joelthelion wrote: post #3

I wonder if you could creatively combine this model with search algorithms to advance the state of the art in computer chess? I wouldn't be surprised to see such a bot pop up on tcec in a couple years.

The thing is classical chess (unlike eg; go) is essentially "solved" when run on computers capable of extreme depth. Modern chess engines play essentially flawlessly.

This is accurate for endgames only. In complicated positions, there is still room for improvement - the recent game of lc0 vs stockfish where lc0 forced a draw against an impending checkmate is a good example. There is currently no way for a chess engine searching a massive game tree can see how an innocuous pawn move enables a forced stalemate 40 moves down the line.

Re: Grandmaster-level chess without search

#15

Post by janalsncm »

hlfshell wrote: post #2

I did a talk about this! (And also wrote up about my talk here[1]). This paper is a great example of both knowledge distillation. It's less of a paper about chess and more about how complicated non linear search functions - complete with whatever tuning experts can prepare - can be distilled into a (quasi-linear, if it's a standardized input like chess) transformer model. [1]: https://hlfshell.ai/posts/deepmind-grand…

I think the vs. humans result should be taken with a huge grain of salt. These are blitz games, and their engine’s elo was far higher against humans than against other bots. So it’s likely that time was a factor, where humans are likely to flag (run out of time) or blunder in low time situations.

It’s still very cool that they could learn a very good eval function that doesn’t require search. I would’ve liked the authors to throw out the games where the Stockfish fallback kicked in though. Even for a human, mate in 2 vs mate in 10 is the difference between a win and a draw/loss on time.

I also would’ve liked to see a head to head with limited search depth Stockfish. That would tell us approximately how much of the search tree their eval function distilled.

Re: Grandmaster-level chess without search

#16

Post by hlfshell »

janalsncm wrote: post #15
hlfshell wrote: post #2

I did a talk about this! (And also wrote up about my talk here[1]). This paper is a great example of both knowledge distillation. It's less of a paper about chess and more about how complicated non linear search functions - complete with whatever tuning experts can prepare - can be distilled into a (quasi-linear, if it's a standardized input like chess) transformer model. [1]: https://hlfshell.ai/posts/deepmind-grand…

I think the vs. humans result should be taken with a huge grain of salt. These are blitz games, and their engine’s elo was far higher against humans than against other bots. So it’s likely that time was a factor, where humans are likely to flag (run out of time) or blunder in low time situations. It’s still very cool that they could learn a very good eval function that doesn’t require search. I would’ve liked the aut…

The reason the time (blitz) games make sense is because the distilled functionality is of a 50ms Stockfish eval function. The engine likely would perform worse as only the human would benefit from the additional time.

As for limited search tree I like the idea! I think it's tough to measure, since the time it takes to perform search across various depths vary wildly based on the complexity of the position. I feel like you would have to compile a dataset of specific positions identified to require significant depth of search to find a "good" move.

Re: Grandmaster-level chess without search

#17

Post by KK7NIL »

GaggiX wrote: post #10
KK7NIL wrote: post #6

Earlier quoted context omitted.

The developers of stockfish and lc0 (and the many weaker engines around) would disagree, we've seen their strength improve considerably over the last few years. Currently there's a very interesting war between small neural networks on the CPU with high search depth alpha-beta pruning (stockfish NNUE) and big neural networks on a GPU with Monte Carlo search and lower depth (lc0). So, while machines beating humans is "…

Stockfish and lc0 would always draw if they are not put in unbalanced starting positions, the starting position will be swapped in the next game to make it fair.

In classical controls (what TCEC mainly uses), yes. They can play pretty exciting bullet chess without a forced opening though.

Re: Grandmaster-level chess without search

#18

Post by janalsncm »

solveit wrote: post #7
alfalfasprout wrote: post #5

Earlier quoted context omitted.

The thing is classical chess (unlike eg; go) is essentially "solved" when run on computers capable of extreme depth. Modern chess engines play essentially flawlessly.

We really have no way to know this. But I would be very surprised if modern chess engines didn't regularly blunder into losing (from the perspective of a hypothetical 32-piece tablebase) positions, and very very surprised if modern chess engines perfectly converted tablebase-winning positions.

The fact that TCEC games aren’t all draws suggests that computers aren’t perfect. Stockfish loses to Leela sometimes for example.

Re: Grandmaster-level chess without search

#19

Post by janalsncm »

hlfshell wrote: post #16
janalsncm wrote: post #15

Earlier quoted context omitted.

I think the vs. humans result should be taken with a huge grain of salt. These are blitz games, and their engine’s elo was far higher against humans than against other bots. So it’s likely that time was a factor, where humans are likely to flag (run out of time) or blunder in low time situations. It’s still very cool that they could learn a very good eval function that doesn’t require search. I would’ve liked the aut…

The reason the time (blitz) games make sense is because the distilled functionality is of a 50ms Stockfish eval function. The engine likely would perform worse as only the human would benefit from the additional time. As for limited search tree I like the idea! I think it's tough to measure, since the time it takes to perform search across various depths vary wildly based on the complexity of the position. I feel lik…

My point is that if the computer never flags it will have an inherent advantage in low time controls. If not, why not just test it in hyperbullet games? Games where humans flag in a drawn or winning position need to be excluded, otherwise it’s unclear what this is even measuring.

And limited depth games would not have been difficult to run. You can run a limited search Stockfish on a laptop using the UCI protocol: https://github.com/official-stockfish/Stockfish/wiki/UCI-%26...

Re: Grandmaster-level chess without search

#20

Post by tzs »

OT: what's the state of the art in non-GM level computer chess?

Say I want to play chess with an opponent that is at about the same skill level as me, or perhaps I want to play with an opponent about 100 rating points above me for training.

Most engines let you dumb them down by cutting search depth, but that usually doesn't work well. Sure, you end up beating them about half the time if you cut the search down enough but it generally feels like they were still outplaying you for much of the game and you won because they made one or two blunders.

What I want is a computer opponent that plays at a level of my choosing but plays a game that feels like that of a typical human player of that level.

Are there such engines?

Post reply on HN