Skip to yearly menu bar Skip to main content


Poster

Accelerated Algorithms for Smooth Convex-Concave Minimax Problems with O(1/k^2) Rate on Squared Gradient Norm

TaeHo Yoon · Ernest Ryu

Keywords: [ Convex Optimization ]


Abstract: In this work, we study the computational complexity of reducing the squared gradient magnitude for smooth minimax optimization problems. First, we present algorithms with accelerated O(1/k2) last-iterate rates, faster than the existing O(1/k) or slower rates for extragradient, Popov, and gradient descent with anchoring. The acceleration mechanism combines extragradient steps with anchoring and is distinct from Nesterov's acceleration. We then establish optimality of the O(1/k2) rate through a matching lower bound.

Chat is not available.