2025
Optimism Without Regularization: Constant Regret in Zero-Sum Games
NeurIPS 2025poster
This paper studies the *optimistic* variant of Fictitious Play for learning in two-player zero-sum games. While it is known that Optimistic FTRL -- a *regularized* algorithm with a bounded stepsize parameter -- obtains constant regret in this setting, we show for the first time that similar, optima…