Papers
Topics
Authors
Recent
2000 character limit reached

Local convergence of simultaneous min-max algorithms to differential equilibrium on Riemannian manifold (2405.13392v3)

Published 22 May 2024 in cs.LG, math.OC, and stat.ML

Abstract: We study min-max algorithms to solve zero-sum differential games on Riemannian manifold. Based on the notions of differential Stackelberg equilibrium and differential Nash equilibrium on Riemannian manifold, we analyze the local convergence of two representative deterministic simultaneous algorithms $\tau$-GDA and $\tau$-SGA to such equilibria. Sufficient conditions are obtained to establish the linear convergence rate of $\tau$-GDA based on the Ostrowski theorem on manifold and spectral analysis. To avoid strong rotational dynamics in $\tau$-GDA, $\tau$-SGA is extended from the symplectic gradient-adjustment method in Euclidean space. We analyze an asymptotic approximation of $\tau$-SGA when the learning rate ratio $\tau$ is big. In some cases, it can achieve a faster convergence rate to differential Stackelberg equilibrium compared to $\tau$-GDA. We show numerically how the insights obtained from the convergence analysis may improve the training of orthogonal Wasserstein GANs using stochastic $\tau$-GDA and $\tau$-SGA on simple benchmarks.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (41)
  1. Optimization Algorithms on Matrix Manifolds. Princeton University Press, Princeton, 2008.
  2. Existence, stability and scalability of orthogonal convolutional neural networks. Journal of Machine Learning Research, 23:1–56, jan 2022.
  3. A Tight and Unified Analysis of Gradient-Based Methods for a Whole Spectrum of Differentiable Games. In International Conference on Artificial Intelligence and Statistics, volume 108 of Proceedings of Machine Learning Research, pages 2863–2873, Virtual Event, 2020.
  4. The Mechanics of n-Player Differentiable Games. In International Conference on Machine Learning, volume 80, pages 354–363, Stockholmsmässan, Stockholm Sweden, 2018.
  5. Some theoretical insights into Wasserstein GANs. Journal of Machine Learning Research, 22:1–45, 2021.
  6. Smooth Maximum Unit: Smooth Activation Function for Deep Networks using Smoothing Maximum Technique. In Conference on Computer Vision and Pattern Recognition, pages 784–793, New Orleans, LA, USA, 2022.
  7. Nicolas Boumal. An Introduction to Optimization on Smooth Manifolds. Cambridge University Press, 2023.
  8. J Bruna and S Mallat. Invariant Scattering Convolution Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(8):1872–1886, 2013.
  9. Parseval networks: improving robustness to adversarial examples. In International Conference on Machine Learning, pages 854–863, Sydney, Australia, 2017.
  10. The Limit Points of (Optimistic) Gradient Descent in Min-Max Optimization. In Advances in neural information processing systems, pages 9256–9266, Red Hook, NY, USA, 2018.
  11. Optimistic Gradient Descent Ascent in Zero-Sum and General-Sum Bilinear Games. arXiv preprint arXiv:2208.03085, 2022.
  12. Local convergence analysis of gradient descent ascent with finite timescale separation. In International Conference on Learning Representation, 2021.
  13. Implicit learning dynamics in stackelberg games: Equilibria characterization, convergence analysis, and empirical study. In International Conference on Machine Learning, pages 3133–3144, Virtual Event, 2020. PMLR.
  14. Global Convergence to the Equilibrium of GANs using Variational Inequalities. arXiv preprint arXiv:1808.01531, 2018.
  15. A Variational Inequality Perspective on Generative Adversarial Networks. In International Conference on Learning Representations, New Orleans, LA, USA, 2019.
  16. Improved training of wasserstein gans. In Advances in neural information processing systems, volume 30, pages 5767–5777, Long Beach, CA, USA, 2017.
  17. Riemannian Hamiltonian methods for min-max optimization on manifolds. SIAM Journal on Optimization, 33(3):1797–1827, 2023.
  18. The Limits of Min-Max Optimization Algorithms: Convergence to Spurious Non-Critical Sets. In International Conference on Machine Learning, volume 139, pages 4337–4348, Virtual Event, 2021.
  19. Extragradient Type Methods for Riemannian Variational Inequality Problems. arXiv preprint arXiv:2309.14155, 2023.
  20. Gradient Descent Ascent for Minimax Problems on Riemannian Manifolds. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1(7):8466 – 8476, 2023.
  21. What is local optimality in nonconvex-nonconcave minimax optimization? In International Conference on Machine Learning, pages 4880–4889, Virtual Event, 2020.
  22. First-Order Algorithms for Min-Max Optimization in Geodesic Metric Spaces. In Advances in Neural Information Processing Systems, volume 35, pages 6557–6574, New Orleans, Louisiana, USA, 2022.
  23. Differentiable Game Mechanics. Journal of Machine Learning Research, 20(84):1–40, 2019.
  24. On convergence of gradient descent ascent: A tight local analysis. In International Conference on Machine Learning, pages 12717–12740, Baltimore, Maryland USA, 2022. PMLR.
  25. On Gradient Descent Ascent for Nonconvex-Concave Minimax Problems. In International Conference on Machine Learning, pages 6083–6093, Virtual Event, 2020.
  26. Stochastic Hamiltonian Gradient Methods for Smooth Games. In International Conference on Machine Learning, volume 119, pages 6370–6381, Virtual Event, 2020.
  27. Tight analysis of extra-gradient and optimistic gradient methods for nonconvex minimax problems. Advances in Neural Information Processing Systems, 35:31213–31225, 2022.
  28. Jonathan H Manton. Optimization algorithms exploiting unitary constraints. IEEE Transactions on Signal Processing, 50(3):635–650, 2002.
  29. Optimistic mirror descent in saddle-point problems: Going the extra (gradient) mile. In International Conference on Learning Representations, New Orleans, LA, USA, 2019.
  30. Orthogonal wasserstein gans. arXiv preprint arXiv:1911.13060, 2019.
  31. Iterative Solution of Nonlinear Equations in Several Variables. New York : Academic Press, 1970.
  32. Scaling the Scattering Transform: Deep Hybrid Networks. In International Conference on Computer Vision, Venice, Italy, 2017.
  33. Nonconvex min-max optimization: Applications, challenges, and recent theoretical advances. IEEE Signal Processing Magazine, 37(5):55–66, 2020.
  34. The earth mover’s distance as a metric for image retrieval. International journal of computer vision, 40:99–121, 2000.
  35. W Rudin. Principles of Mathematical Analysis. McGraw-Hill Publishing Company, 3rd edition, 1976.
  36. Maximilian Seitzer. pytorch-fid: FID Score for PyTorch. https://github.com/mseitzer/pytorch-fid, 2020.
  37. Riemannian Optimistic Algorithms. arXiv preprint arXiv:2308.16004, 2023.
  38. On Solving Minimax Optimization Locally: A Follow-the-Ridge Approach. In International Conference on Learning Representations, Addis Ababa, Ethiopia, 2020.
  39. Xingzhi Zhan. Matrix Inequalities. Springer Berlin, Heidelberg, 2002.
  40. Near-optimal local convergence of alternating gradient descent-ascent for minimax optimization. In International Conference on Artificial Intelligence and Statistics, pages 7659–7679, Virtual Event, 2022.
  41. Sion’s Minimax Theorem in Geodesic Metric Spaces and a Riemannian Extragradient Algorithm. SIAM Journal on Optimization, 33(4), 2023.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now

Whiteboard

Paper to Video (Beta)

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now

Continue Learning

We haven't generated follow-up questions for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now

Authors (1)

  1. Sixin Zhang

Collections

Sign up for free to add this paper to one or more collections.

User Circle Single Streamline Icon: https://streamlinehq.com Sign Up

Tweets

Sign up for free to view the 4 tweets with 1 like about this paper.

User Circle Single Streamline Icon: https://streamlinehq.com Sign Up for Free