2019/07/11 by Wolfgang Konen, Konen, Wolfgang
Computer Science · Psychology · #Artificial Intelligence in Games #Educational Games and Gamification #Reinforcement Learning in Robotics
paper · pdf · doi:10.48550/arxiv.1907.06508
We present a new general board game (GBG) playing and learning framework. GBG\ndefines the common interfaces for board games, game states and their AI agents.\nIt allows one to run competitions of different agents on different games. It\nstandardizes those parts of board game playing and learning that otherwise\nwould be tedious and repetitive parts in coding. GBG is suitable for arbitrary\n1-, 2-, ..., N-player board games. It makes a generic TD(\λ)-n-tuple\nagent for the first time available to arbitrary games. On various games,\nTD(\λ)-n-tuple is found to be superior to other generic agents like\nMCTS. GBG aims at the educational perspective, where it helps students to start\nfaster in the area of game learning. GBG aims as well at the research\nperspective by collecting a growing set of games and AI agents to assess their\nstrengths and generalization capabilities in meaningful competitions. Initial\nsuccessful educational and research results are reported.\n