Variance-reduced accelerated methods for decentralized stochastic double-regularized nonconvex strongly-concave minimax problems
math.OC, cs.LG
Submitted: 2023-07-14
Updated: 2026-10-02
Code: https://github.com/RPI-OPT/VRLM
Terminology
Sources
- DAMA: A Unified Accelerated Approach for Decentralized Nonconvex Minimax Optimization-Part I: Algorithm Development and Results
- An Efficient Stochastic Algorithm for Decentralized Nonconvex-Strongly-Concave Minimax Optimization
- Accelerated Proximal Alternating Gradient-Descent-Ascent for Nonconvex Minimax Machine Learning
- Decentralized Stochastic Gradient Descent Ascent for Finite-Sum Minimax Problems
- A variance reduced framework for (non)smooth nonconvex-nonconcave stochastic minimax problems with extended Kurdyka-Lojasiewicz property
- Gradient Tracking: A Unified Approach to Smooth Distributed Optimization
- ZeroSARAH: Efficient Nonconvex Finite-Sum Optimization with Zero Full Gradient Computation
- Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
- A Stochastic Proximal Gradient Framework for Decentralized Non-Convex Composite Optimization: Topology-Independent Sample Complexity and Communication Efficiency
- Universal Gradient Descent Ascent Method for Nonconvex-Nonconcave Minimax Optimization
Related papers
- Lions and Muons: Optimization via Stochastic Frank-Wolfe under Heavy-Tailed Noise
- Adam-HNAG: A Convergent Reformulation of Adam with Accelerated Rate
- Incremental Learning in Mirror Flows
- Second-Order Optimization with Asynchronous Curvature
- Online Control via Counterfactual Tracking
- Asynchronous Replanning in Two Population Linear Quadratic Mean Field Games: Information Requirements and Stability