It's very interesting to see if it is able to handle much more advanced and tuned engines that exist for chess, game with considerable much more complicated rules?
It'd be particularly useful to have a chess bot that can play badly in the same way a human does.
The problem with the current chess bots is that they play badly, badly. They choose a terrible random mistake to make every few moves, while some of their other moves are brilliant. They cannot accurately mimic beginner or intermediate level players.
This seems like something DeepMind could create, given the incentive. They were able to train AlphaGo to predict human moves in Go at a very high accuracy (obviously not with AlphaGo Zero, but the inferior human-predictive version is how they determined that AGZ is playing qualitatively differently).
I have some idea how it MIGHT work, but it would be a very boring solution involving 'learning' Stockfish's parameters and HOPING to find improvements to something like integrating time management and search/pruning into it.
I wouldn't bet on it though. SMP is notoriously hard to work with alpha-beta search and there are a lot of clever tricks (which is probably still not perfect). Maybe with ASICs, you could make it stronger, but then it wouldn't be as fair a comparison.
Well, all top engines did some kind of search on parameters, not sure if you can find much improvement there.
I'm talking about something similar to the described in the paper, 100% self-learned solution without using human heuristics, based on NNs. That could bring a totally new ideas into chess.
Shogi is probably the closest historical game in terms of complexity to Go. Some of the larger variants might exceed Go's complexity if played with drops, though that's not normally done. And Go played on a 9x9 board (like standard Shogi) has a substantially lower state space complexity (and almost certainly lower by other measures as well.)
But shogi is much more obscure outside of Japan than go or chess, so it gets less interest, especially in the large-board variants.
I think that the existence of highly optimized chess AI makes it interesting from two angles:
1) Generalization: Can one make AI using same approach that can play both chess and Go at superhuman levels
2) Efficiency: Can these newer methods match or outperform also in terms of compute/energy costs
But maybe not sexy enough, or we just don't hear about it as much.
That makes it even more interesting. I think it would be very notable and significant if a neural network with MCTS and self-play reinforcement learning could surpass Stockfish, which has superhuman strength but was developed with an utterly different approach involving lots of human guidance and grandmaster input.
Giraffe attempted this (with more standard tree search than MCTS and with only a value function rather than a combined policy/value network), but only reached IM level -- certainly impressive, but nowhere close to Stockfish.
Denis Hassabis was asked this in a Q&A after a talk he gave and according to him someone did this (bootstrap a chess engine from self play) successfully, while still being a student and was hired by them subsequently.
I didn't see the talk, but I'm guessing he was referring to the Giraffe engine done by Matthew Lai (https://arxiv.org/abs/1509.01549). The main thing there is that he only learns an evaluation function, not a policy. Giraffe still uses classical alpha-beta search over the full action space. AFAIK nobody has learned a decent policy network for chess, probably because 1) it's super tactical, and 2) nobody cares that much because alpha-beta is so strong
Minimax with Alpha Beta pruning works in Chess because the search tree is way smaller. The reason why all this "Monte-Carlo Tree Search + Neural Nets" are being used in Go because Minimax + Alpha Beta pruning DOESN'T work in Go.
It's very interesting to see if it is able to handle much more advanced and tuned engines that exist for chess, game with considerable much more complicated rules?