The improvement is due to the embedded neural network providing much more accurate evaluation and also in some sense gaining an extra ply or so of search by “seeing” some tactics that a normal eval won’t spot. With four threads, the gains were 156 elo in MCTS and 170 in standard mode. The reinforcement learning phase for Dragon is in its infancy, but is already showing great promise.ĭragon is a huge strength improvement over 14.1, the last release, about 197 elo in MCTS mode and 189 elo in standard mode, at CCRL blitz time control on one thread based on direct matches. Training an NNUE network based on this evaluation was both an advantage and a challenge, requiring experimentation with architectures and data generation of billions of positions. Komodo has a great deal of chess knowledge in its evaluation. Dragon uses NNUE (Neural Network Updated Efficiently) technology, originally developed for the game of shogi.
0 Comments
Leave a Reply. |