A survey of planning and learning in games

摘要: In general, games pose interesting and complex problems for the implementation of intelligent agents are a popular domain in study artificial intelligence. fact, have been at center some most well-known achievements From classical board such as chess, checkers, backgammon Go, to video Dota 2 StarCraft II, intelligence research has devised computer programs that can play level human master even world champion level. Planning learning, two successful paradigms intelligence, greatly contributed these achievements. Although representing distinct approaches, planning learning try solve similar share similarities. They complement each other. This led on methodologies combine strengths both approaches derive better solutions. paper presents survey multiple proposed integrate context games. order provide richer contextualization, also techniques commonly used games, terms their theoretical foundations applications.