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

Grandmaster-Level Chess Without Search

arxiv.org

Post reply on HN
Print view View original

81–90 of 133 posts

Re: Grandmaster-Level Chess Without Search

#81

Post by salamo »

sp332 wrote: post #75
salamo wrote: post #15

I think this is an interesting finding from a practical perspective. A function which can reliably approximate stockfish at a certain depth could replace it, basically "compressing" search to a set depth. And unlike NNUE which is optimized for CPU, a neural network is highly parallelizable on GPU meaning you could send all possible future positions (at depth N) through the network and use the results for a primitive…

The Stockfish installer is ~45 MB. At 16 bits per parameter, the 270B model would be over 500 MB. The 9B model would be smaller than Stockfish, but you could probably find a smaller chess engine that achieves 2000 ELO.

The advantage of this approach that we can run many simultaneous computations on the GPU/TPU. Instead of using maybe a dozen CPU threads, we can approximate the value of a few thousand positions at the same time.

Re: Grandmaster-Level Chess Without Search

#82

Post by bigmadshoe »

spencerchubb wrote: post #14
xianshou wrote: post #10

The path to AGI: 0. Have model A. 1. Use Monte Carlo with A to get supervised data. 2. Train model B with data from A. 3. Use Monte Carlo with B to get supervised data. 4. Train model C with data from B...

That is an awesome idea. I wish the authors would open source the code and weights so this can be tried.

They basically just described alphaZero, the difference being that alphaZero uses MCTS during inference too.

Re: Grandmaster-Level Chess Without Search

#83

Post by GavinB »

janalsncm wrote: post #29
paulddraper wrote: post #18

There is rampant misunderstanding of some parts of this article; allow me to help :) The "no-search" chess engine uses search (Stockfish) in in two ways: 1. To score positions in the training data. This is only training data, no search is performed when actually playing. 2. To play moves when the position has many options with a 99% win rate. This is to prevent pathological behavior in already won positions, and is n…

In one sense, I can understand why they would choose to use Stockfish in mate-in-N positions. The fact that the model can't distinguish between mate in 5 and mate in 3 is an implementation detail. Since the vast majority of positions are not known to be wins or draws, it's still an interesting finding. However, in reality all positions are actually wins (for black or white) or draws. One reason they gave for why stoc…

Is this in comparison to some other evaluation function which is perfect? I agree that all positions should have a certainty of win, draw, or lose with perfect play, but no engine is close to that level of evaluation function.

I do suspect that this pathological behavior could be trained out with additional fine tuning, but likely not without slightly diminishing the model's overall ability.

Re: Grandmaster-Level Chess Without Search

#85

Post by rkagerer »

YeGoblynQueenne wrote: post #51
paulddraper wrote: post #18

There is rampant misunderstanding of some parts of this article; allow me to help :) The "no-search" chess engine uses search (Stockfish) in in two ways: 1. To score positions in the training data. This is only training data, no search is performed when actually playing. 2. To play moves when the position has many options with a 99% win rate. This is to prevent pathological behavior in already won positions, and is n…

>> 1. To score positions in the training data. This is only training data, no search is performed when actually playing. That's like saying you can have eggs without chickens, because when you make an omelette you don't add chickens. It's completely meaningless and a big fat lie to boot. The truth is that the system created by DeepMind consists of two components: a search-based system used to annotate a dataset of mo…

Ok maybe it's fair to say it's Chess with Precompiled Search.

Re: Grandmaster-Level Chess Without Search

#86

Post by lettergram »

I immediately saw this and knew it was BS. It’s a search problem, the human brain even does a search. Model internally is scanning each position and determining the next probably position. That is a predictive search, you can’t just restructure the problem.

Now arguably it’s doing it differently, maybe? But still a search

Re: Grandmaster-Level Chess Without Search

#87

Post by porphyra »

fiforpg wrote: post #2

Given that they used position evaluation from (a search chess engine[1]) Stockfish, how is this "without search"? Edit: looking further than the abstract, this is rather an exploration of scale necessary for a strong engine. Could go without "without search" in the title I guess. [1]: IIRC, it also uses a Leela-inspired NN for evaluation.

Does Stockfish really use a Leela-inspired NN? I thought the NNUE was independently developed and completely different (it's a very tiny network that runs on the CPU).

Re: Grandmaster-Level Chess Without Search

#88

Post by rolisz »

joe_the_user wrote: post #44
janalsncm wrote: post #40

Earlier quoted context omitted.

A sufficiently large nn can learn an arbitrary function, yes. But stockfish is also theoretically perfect given infinite computational resources. What is interesting is performing well under reasonable computational constraints i.e. doing it faster/with fewer flops than stockfish.

Is the model more efficient than Stockfish? I think Stockfish runs on regular CPU computer and I'd guess this " 270M parameter transformer model" requires a GPU but I can't find any reference to efficiency in the paper. Also found in the paper: "While our largest model achieves very good performance, it does not completely close the gap to Stockfish 16". It's actually inferior but they still think it's an interesting…

BERT has around that many parameters and it runs on CPU in 200ms

Re: Grandmaster-Level Chess Without Search

#89

Post by ad8e »

A quote from discord: "apparently alpha-zero has been replicated in open source as leela-zero, and then leela-zero got a bunch of improvements so it's far ahead of alpha-zero. but leela-zero was barely mentioned at all in the paper; it was only dismissed in the introduction and not compared in the benchmarks. in the stockfish discord they are saying that leela zero can already do everything in this paper including using the transformer architecture."

Re: Grandmaster-Level Chess Without Search

#90

Post by Oreb »

porphyra wrote: post #87
fiforpg wrote: post #2

Given that they used position evaluation from (a search chess engine[1]) Stockfish, how is this "without search"? Edit: looking further than the abstract, this is rather an exploration of scale necessary for a strong engine. Could go without "without search" in the title I guess. [1]: IIRC, it also uses a Leela-inspired NN for evaluation.

Does Stockfish really use a Leela-inspired NN? I thought the NNUE was independently developed and completely different (it's a very tiny network that runs on the CPU).

This is true, but at least for a while (I’m not sure if it’s still the case), Leela data was used (along with data generated from Stockfish self-play) to train Stockfish’s NN.
Post reply on HN