You're looking at Less Wrong's discussion board. This includes all posts, including those that haven't been promoted to the front page yet. For more information, see About Less Wrong.

ShardPhoenix comments on [LINK] Deep Learning Machine Teaches Itself Chess in 72 Hours - Less Wrong Discussion

8 Post author: ESRogs 14 September 2015 07:38PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Article Navigation

Comments (15)

Comments (15)

You are viewing a single comment's thread. Show more comments above.

Comment author: ShardPhoenix 14 September 2015 11:34:48PM *  9 points [-]

Although it's not better than existing solutions, it's a cool example of how good results can be achieved in a relatively automatic way - by contrast, the evaluation functions of the best chess engines have been carefully engineered and fine-tuned over many years, at least sometimes with assistance from people who are themselves master-level chess players. On the other hand this neural network approach took a relatively short time and could have been applied by someone with little chess skill.

edit: Reading the actual paper, it does sound like a certain amount of iteration, and expertise on the author's part, was still required.

edit2: BTW, the paper is very clear and well written. I'd recommend giving it a read if you're interested in the subject matter.

Comment author: TheMajor 15 September 2015 07:08:02PM 2 points [-]

Thank you for recommending to read the paper, I don't think I would have otherwise and I greatly enjoyed reading it!

Comment author: V_V 15 September 2015 09:07:26PM -1 points [-]

Although it's not better than existing solutions, it's a cool example of how good results can be achieved in a relatively automatic way - by contrast, the evaluation functions of the best chess engines have been carefully engineered and fine-tuned over many years, at least sometimes with assistance from people who are themselves master-level chess players. On the other hand this neural network approach took a relatively short time and could have been applied by someone with little chess skill.

But how much of its performance comes from the neural network learning some non-trivial evaluation function and how much comes from brute-forcing the game tree on a modern computer?

If the neural network was replaced by a trivial heuristic, say, material balance, how would the engine perform?

Comment author: ShardPhoenix 15 September 2015 11:31:19PM *  2 points [-]

In the paper they start with just material balance - then via the learning process, their score on the evaluation test goes from "worse than all hand-written chess engines" to "better than all except the very best one" (and the best one, while more hand-crafted, also uses some ML/statistical tuning of numeric params, and has had a lot more effort put into it).

The reason why the NN solution currently doesn't do as well in real games is because it's slower to evaluate and therefore can't brute-force as far.

Powered by Reddit Powered by Reddit