The Stockfish chess engine has switched to a new evaluation method similar to what Alpha Zero or Lela Chess uses. It is no longer about material but winning chances, which is dependent upon an AI that can train itself. This method allowed the program Alpha Go to beat the best Go players, which was previously thought to be impossible.
"New evaluation
This release also introduces a new convention for the evaluation that is reported by search. An evaluation of +1 is now no longer tied to the value of one pawn, but to the likelihood of winning the game. With a +1 evaluation, Stockfish has now a 50% chance of winning the game against an equally strong opponent. This convention scales down evaluations a bit compared to Stockfish 15 and allows for consistent evaluations in the future."
https://stockfishchess.org/blog/2022/stockfish-15-1/
This release also introduces a new convention for the evaluation that is reported by search. An evaluation of +1 is now no longer tied to the value of one pawn, but to the likelihood of winning the game. With a +1 evaluation, Stockfish has now a 50% chance of winning the game against an equally strong opponent. This convention scales down evaluations a bit compared to Stockfish 15 and allows for consistent evaluations in the future."
https://stockfishchess.org/blog/2022/stockfish-15-1/
No comments:
Post a Comment