Hi Remi, hi friends, > For the moment, my main objective is shogi. I > will participate in the World Computer Shogi > Championship in May.
Good luck! Please, keep us informed when the tournament is running. > So I am developing a game-independent AlphaZero framework. I am hoping several people are working in this direction. Explanation: In 1998 "Zillions of Games" (by Mark Lefler and Jeff Mallet) was made public. It is a multi-game program. The user only has to formulate the rules of a game in a script language (zrf), and then the engine is playing this game, based on alpha-beta with an evaluation function where each piece gets its (local+global) mobility assigned. (Mark Lefler is co-programmer of current top chess engine Komodo.) In 2008, Cameron Browne published his Ph.D. thesis on "automated game design", where a new game is evaluated by selfplay of an alpha-beta engine. It would be fantastic to have new programs which learn to play newly designed games in AlphaZero mode. This would revolutionize the design of new board games. Side question: Is someone investigating AlphaZero approaches for non-zero-sum games of for games with more than two players? Ingo. _______________________________________________ Computer-go mailing list Computer-go@computer-go.org http://computer-go.org/mailman/listinfo/computer-go