Next-Generation Iterative Algorithms for Large-Scale Min-Max Optimization: Design and Analysis

TL;DR

This paper introduces innovative iterative algorithms for large-scale min-max optimization, providing new theoretical insights and practical solutions to improve performance in machine learning, game theory, and optimization applications.

Contribution

It presents novel algorithms and theoretical analysis that overcome limitations of existing methods for large-scale min-max problems.

Findings

New algorithms with improved convergence properties

Theoretical guarantees for algorithm performance

Enhanced applicability to real-world problems

Abstract

This thesis investigates the design of algorithms for solving min-max optimization problems, which form the mathematical foundation of many modern applications in machine learning, game theory, and optimization. This work offers new theoretical insights and practical algorithms that address the limitations of existing methods in various problem settings.