mlsanityData to 14 Oct 2026

Stabilizing GANs' Training with Brownian Motion Controller

arXiv 2306.10468 ↗First posted 18 Jun 2023cs.LG math.DScited by 4

Not on the map. Why?

Abstract (arXiv)

The training process of generative adversarial networks (GANs) is unstable and does not converge globally. In this paper, we examine the stability of GANs from the perspective of control theory and propose a universal higher-order noise-based controller called Brownian Motion Controller (BMC). Starting with the prototypical case of Dirac-GANs, we design a BMC to retrieve precisely the same but reachable optimal equilibrium. We theoretically prove that the training process of DiracGANs-BMC is globally exponential stable and derive bounds on the rate of convergence. Then we extend our BMC to normal GANs and provide implementation instructions on GANs-BMC. Our experiments show that our GANs-BMC effectively stabilizes GANs' training under StyleGANv2-ada frameworks with a faster rate of convergence, a smaller range of oscillation, and better performance in terms of FID score.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 23 of its 50 references that are in the map. First posted on arXiv 18 Jun 2023.

Builds on

23 of 50 references.

References · 23

  1. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  2. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,588
  3. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,988
  4. 2016Improved Techniques for Training GANscited by 2,970
  5. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,626
  6. 2017Improved Training of Wasserstein GANscited by 2,235
  7. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,054
  8. 2018Spectral Normalization for Generative Adversarial Networkscited by 1,636
  9. 2015LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loopcited by 1,355
  10. 2020Training Generative Adversarial Networks with Limited Datacited by 1,186
  11. 2021Alias-Free Generative Adversarial Networkscited by 1,184
  12. 2018Which Training Methods for GANs do actually Converge?cited by 679
11 more
  1. 2016f-GAN: Training Generative Neural Samplers using Variational Divergence Minimizationcited by 506
  2. 2017Towards Principled Methods for Training Generative Adversarial Networkscited by 464
  3. 2017MMD GAN: Towards Deeper Understanding of Moment Matching Networkcited by 254
  4. 2016Energy-based Generative Adversarial Networkcited by 119
  5. 2017PacGAN: The power of two samples in generative adversarial networkscited by 104
  6. 2016Disentangling factors of variation in deep representations using adversarial trainingcited by 87
  7. 2016Amortised MAP Inference for Image Super-resolutioncited by 68
  8. 2016Generative Multi-Adversarial Networkscited by 60
  9. 2017Many Paths to Equilibrium: GANs Do Not Need to Decrease a Divergence At Every Stepcited by 45
  10. 2021Improving GAN Equilibrium by Raising Spatial Awarenesscited by 22
  11. 2019Understanding and Stabilizing GANs' Training Dynamics with Control Theorycited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 1; 2025: 1; 2026: 1.

4 papers.

  1. 2023Monotone Generative Modeling via a Gromov-Monge Embeddingcited by 3
  2. 2024C-GAIL: Stabilizing Generative Adversarial Imitation Learning with Control Theorycited by 3
  3. 2025Performative Drift Resistant Classification Using Generative Domain Adversarial Networkscited by 1
  4. 2026Controlled-Minimax: Stabilizing Adversarial Optimization via Control Theorycited by 0