SDE approximations of GANs training and its long-run behavior

2020/06/03 by Haoyang Cao, Xin Guo, Cao, Haoyang +1
Computer Science · Engineering · #Advanced Decision-Making Techniques #Blasting Impact and Analysis #FOS: Computer and information sciences #FOS: Mathematics #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Neural Networks and Applications #Probability (math.PR)

paper · pdf · doi:10.48550/arxiv.2006.02047

openalex publication_date 2020/06/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

This paper analyzes the training process of GANs via stochastic differential equations (SDEs). It first establishes SDE approximations for the training of GANs under stochastic gradient algorithms, with precise error bound analysis. It then describes the long-run behavior of GANs training via the invariant measures of its SDE approximations under proper conditions. This work builds theoretical foundation for GANs training and provides analytical tools to study its evolution and stability.

Related

• A Solvable High-Dimensional Model of GAN
• Functional Space Analysis of Local GAN Convergence
• HGAN-SDEs: Learning Neural Stochastic Differential Equations with Hermite-Guided Adversarial Training
• Convergence and Sample Complexity of SGD in GANs
• Convergence of GANs Training: A Game and Stochastic Control Methodology