SDE approximations of GANs training and its long-run behavior
arXiv:2006.02047 · doi:10.1017/jpr.2023.57
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.
References in corpus (14)
- GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
- Semantic Segmentation using Adversarial Networks
- Towards Principled Methods for Training Generative Adversarial Networks
- Quant GANs: Deep Generation of Financial Time Series
- Improving Generalization and Stability of Generative Adversarial Networks
- Stabilizing Generative Adversarial Networks: A Survey
- Deep Learning Theory Review: An Optimal Control and Dynamical Systems Perspective
- Fluctuation-dissipation relations for stochastic gradient descent
- Stochastic Modified Equations and Dynamics of Stochastic Gradient Algorithms I: Mathematical Foundations
- GAN and VAE from an Optimal Transport Point of View
- Implicit competitive regularization in GANs
- Generalization Properties of Optimal Transport GANs with Latent Distribution Learning
- A mean-field analysis of two-player zero-sum games
- Learning who is in the market from time series: market participant discovery through adversarial calibration of multi-agent simulators