Scalable decision-making for games of imperfect information
Design of AtaraxosAtaraxos consists of two interdependent self-play reinforcement learning processes, realized by transformer networks for set-up selection and move selection; a belief network trained on self-play games of the final selection networks; and a test-time search procedure that composes the move and belief networks.Interdependent self-play processesThe foundation of Ataraxos is its self-play reinforcement learning … Read more