登录
洞察 - Federated Minimax Optimization
Efficient Federated Minimax Optimization Using Stochastic Smoothed Gradient Descent Ascent
FESS-GDA, a new algorithm that utilizes smoothing techniques, can be uniformly applied to solve several classes of federated nonconvex minimax problems and achieve new or better analytical convergence results.
Efficient Adaptive Federated Minimax Optimization for Non-Convex Problems
Our AdaFGDA algorithm efficiently addresses distributed non-convex minimax optimization problems by incorporating adaptive learning rates, achieving lower gradient and communication complexities simultaneously.
