Corpus ID: 225094133

Training Generative Adversarial Networks by Solving Ordinary Differential Equations

@article{Qin2020TrainingGA,
  title={Training Generative Adversarial Networks by Solving Ordinary Differential Equations},
  author={Chongli Qin and Yan Wu and Jost Tobias Springenberg and A. Brock and J. Donahue and T. Lillicrap and Pushmeet Kohli},
  journal={ArXiv},
  year={2020},
  volume={abs/2010.15040}
}
  • Chongli Qin, Yan Wu, +4 authors Pushmeet Kohli
  • Published 2020
  • Computer Science, Mathematics
  • ArXiv
  • The instability of Generative Adversarial Network (GAN) training has frequently been attributed to gradient descent. Consequently, recent methods have aimed to tailor the models and training procedures to stabilise the discrete updates. In contrast, we study the continuous-time dynamics induced by GAN training. Both theory and toy experiments suggest that these dynamics are in fact surprisingly stable. From this perspective, we hypothesise that instabilities in training GANs arise from the… CONTINUE READING

    Figures and Tables from this paper

    References

    SHOWING 1-10 OF 33 REFERENCES
    Adversarial Lipschitz Regularization
    • 9
    Generative Adversarial Nets
    • 20,163
    • Highly Influential
    • PDF
    Achieving Robustness in the Wild via Adversarial Mixing With Disentangled Representations
    • Sven Gowal, Chongli Qin, +4 authors P. Kohli
    • Computer Science, Mathematics
    • 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
    • 2020
    • 14
    • PDF
    Improved Training of Wasserstein GANs
    • 3,787
    • PDF
    Gradient descent GAN optimization is locally stable
    • 207
    • Highly Influential
    • PDF
    Global Convergence to the Equilibrium of GANs using Variational Inequalities
    • 26
    • Highly Influential
    • PDF
    The Mechanics of n-Player Differentiable Games
    • 128
    • PDF
    Which Training Methods for GANs do actually Converge?
    • 446
    • PDF
    Towards Principled Methods for Training Generative Adversarial Networks
    • 1,026
    • PDF
    Reducing Noise in GAN Training with Variance Reduced Extragradient
    • 41
    • PDF