Abstract:
We present CLASSQL, a multi-agent model for playing real-time strategy games, where learning and control of our own team’s units is decentralized; each agent uses its own reinforcement learning process to learn and control units of the same class. Coordination between these agents occurs as a result of a common reward function shared by all agents and synergistic relations in a carefully crafted state and action model for each class. We present results of CLASSQL against the built-in AI in a variety of maps using the Wargus real-time strategy game.
DOI:
10.1609/aiide.v9i1.12667