SentiMATE: Learning to play Chess through Natural Language Processing · HackerPapers discussions

SentiMATE: Learning to play Chess through Natural Language Processing — mentioned by cbutner

Wed, 06 Oct 2021 01:25:27 GMT

SentiMATE[1] looks at one of the reverse problems in a way - training an engine on commentary data - although it's not exactly what you're talking about. I think this line of thinking could eventually lead to automated metrics for commentary evaluation, which could in turn lead to better methods than top-k/top-p for turning a bunch of sequential logits into a sentence or paragraph - basically treat it like MCTS/PUCT also. The problem is that if you look at high-level commentary - maybe Radjabov-MVL on https://www.chess.com/news/view/2021-champions-chess-tour-fi... (I'm not the best judge…

Paper and discussion history. A mention is not an endorsement.

SentiMATE: Learning to play Chess through Natural Language Processing — mentioned by Deimorz

Fri, 02 Aug 2019 01:58:14 GMT

It's a neat idea, but saying that this AI "mastered" chess is a huge exaggeration. In the actual paper [1], it says that it lost 19/100 games against an opponent playing completely random moves. And by "lost", they mean that it was in a worse position after 40 moves. It sounds like the AI isn't even able to consistently finish games, so they just had to try to judge them at that point. [1]: https://arxiv.org/pdf/1907.08321v1.pdf

Paper and discussion history. A mention is not an endorsement.

SentiMATE: Learning to play Chess through Natural Language Processing — HN submission

Mon, 22 Jul 2019 18:37:06 GMT

Deep Learning model learns to play Chess from text-based commentary

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.