Agents
Learning to Play Sequential Games versus Unknown Opponents
To this end, we use kernel-based regularity assumptions to capture and exploit the structure in the opponent's response. We propose a novel algorithm for the learner when playing against an adversarial sequence of opponents.
No-Regret Learning Dynamics for Extensive-Form Correlated Equilibrium Andrea Celli
Specifically, it has been known for more than 20 years that when all players seek to minimize their internal regret in a repeated normal-form game, the empirical frequency of play converges to a normal-form correlated equilibrium.
Counterfactual Conservative Q Learning for Offline Multi-agent Reinforcement Learning Jianzhun Shao, Y un Qu
MARL in real scenarios is still challenging due to the same safety and efficiency concerns in single-agent setting, then it is worth conducting investigation for offline RL in multi-agent setting.