Why would a chess engine falsely assess a position?
Sort:
Most Recent
Forum Legend
Following
New Comments
Locked Topic
Pinned Topic
I just wrote an article about how AlphaZero was able to prove Stockfish's evaluations of positions to wrong on numerous occasions in their match. However, I wasn't really able to come to a definitive conclusion as to why such a powerful engine would get stuff that seems obvious to humans (after a wile, at least) wrong. I went along with the hypothesis given in Game Changer- that since AlphaZero has no preconceptions about chess it is able to view a given position more flexibly than an engine like Stockfish which has been hand-crafted. Does anyone have anything to add to this?