Last Iterate is Slower than Averaged Iterate in Smooth Convex-Concave Saddle Point Problems

Abstract

In this paper we study the smooth convex-concave saddle point problem. Specifically, we analyze the last iterate convergence properties of the Extragradient (EG) algorithm. It is well known that the ergodic (averaged) iterates of EG converge at a rate of $O(1/T)$ (Nemirovski, 2004). In this paper, we show that the last iterate of EG converges at a rate of $O(1/\sqrt{T})$. To the best of our knowledge, this is the first paper to provide a convergence rate guarantee for the last iterate of EG for the smooth convex-concave saddle point problem. Moreover, we show that this rate is tight by proving a lower bound of $\Omega (1/\sqrt{T})$ for the last iterate. This lower bound therefore shows a quadratic separation of the convergence rates of ergodic and last iterates in smooth convex-concave saddle point problems.

arXiv comment27 pages

References in corpus (14)

• Stochastic Gradient Descent for Non-smooth Optimization: Convergence Results and Optimal Averaging Schemes
• A Variational Inequality Perspective on Generative Adversarial Networks
• SBEED: Convergent Reinforcement Learning with Nonlinear Function Approximation
• Optimistic mirror descent in saddle-point problems: Going the extra (gradient) mile
• The Unusual Effectiveness of Averaging in GAN Training
• Reducing Noise in GAN Training with Variance Reduced Extragradient
• Last-iterate convergence rates for min-max optimization
• Accelerating Smooth Games by Manipulating Spectral Shapes
• On the Iteration Complexity of Oblivious First-Order Optimization Algorithms
• Last-Iterate Convergence: Zero-Sum Games and Constrained Min-Max Optimization
• On Lower and Upper Bounds for Smooth and Strongly Convex Optimization Problems
• Negative Momentum for Improved Game Dynamics
• Convergence of Gradient Methods on Bilinear Zero-Sum Games
• Interaction Matters: A Note on Non-asymptotic Local Convergence of Generative Adversarial Networks

Cited by in corpus (5)

• Accelerated Algorithms for Smooth Convex-Concave Minimax Problems with $\mathcal{O}(1/k^2)$ Rate on Squared Gradient Norm
• Tight last-iterate convergence rates for no-regret learning in multi-player games
• Increasing Iterate Averaging for Solving Saddle-Point Problems
• A New Primal-Dual Algorithm for a Class of Nonlinear Compositional Convex Optimization Problems
• A Geometric Structure of Acceleration and Its Role in Making Gradients Small Fast