Generalized Optimistic Methods for Convex-Concave Saddle Point Problems

This lecture introduces a generalized optimistic framework for solving convex-concave saddle point problems, extending to higher-order methods. The method covers composite settings, complexity bounds, and line search schemes. The instructor discusses the convergence rates, stepsize selection, and the implementation of the algorithm.