Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
129 tokens/sec
GPT-4o
28 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Soft-constrained Schrodinger Bridge: a Stochastic Control Approach (2403.01717v2)

Published 4 Mar 2024 in stat.ML, cs.LG, math.OC, and stat.CO

Abstract: Schr\"{o}dinger bridge can be viewed as a continuous-time stochastic control problem where the goal is to find an optimally controlled diffusion process whose terminal distribution coincides with a pre-specified target distribution. We propose to generalize this problem by allowing the terminal distribution to differ from the target but penalizing the Kullback-Leibler divergence between the two distributions. We call this new control problem soft-constrained Schr\"{o}dinger bridge (SSB). The main contribution of this work is a theoretical derivation of the solution to SSB, which shows that the terminal distribution of the optimally controlled process is a geometric mixture of the target and some other distribution. This result is further extended to a time series setting. One application is the development of robust generative diffusion models. We propose a score matching-based algorithm for sampling from geometric mixtures and showcase its use via a numerical example for the MNIST data set.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (56)
  1. An optimal control perspective on diffusion-based generative modeling. In NeurIPS 2022 Workshop on Score-Based Methods, 2022.
  2. Arne Beurling. An automorphism of product measures. Annals of Mathematics, pages 189–200, 1960.
  3. A Blaquiere. Controllability of a Fokker-Planck equation, the Schrödinger system, and a related stochastic optimal control (revised version). Dynamics and Control, 2(3):235–253, 1992.
  4. Peter J Bushell. Hilbert’s metric and positive contraction mappings in a Banach space. Archive for Rational Mechanics and Analysis, 52:330–338, 1973.
  5. Likelihood training of Schrödinger bridge using forward-backward SDEs theory. In International Conference on Learning Representations, 2021a.
  6. Entropic and displacement interpolation: a computational approach using the Hilbert metric. SIAM Journal on Applied Mathematics, 76(6):2375–2396, 2016a.
  7. On the relation between optimal transport and Schrödinger bridges: A stochastic control viewpoint. Journal of Optimization Theory and Applications, 169:671–691, 2016b.
  8. Relaxed Schrödinger bridges and robust network routing. IEEE Transactions on Control of Network Systems, 7(2):923–931, 2019.
  9. Optimal transport in systems and control. Annual Review of Control, Robotics, and Autonomous Systems, 4:89–113, 2021b.
  10. Stochastic control liaisons: Richard Sinkhorn meets Gaspard Monge on a Schrodinger bridge. SIAM Review, 63(2):249–313, 2021c.
  11. Paolo Dai Pra. A stochastic control approach to reciprocal diffusion processes. Applied Mathematics and Optimization, 23(1):313–329, 1991.
  12. Diffusion Schrödinger bridge with applications to score-based generative modeling. Advances in Neural Information Processing Systems, 34:17695–17709, 2021.
  13. Quantitative uniform stability of the iterative proportional fitting procedure. The Annals of Applied Probability, 34(1A):501–516, 2024.
  14. On a least squares adjustment of a sampled frequency table when the expected marginal totals are known. The Annals of Mathematical Statistics, 11(4):427–444, 1940.
  15. Li Deng. The MNIST database of handwritten digit images for machine learning research. IEEE Signal Processing Magazine, 29(6):141–142, 2012.
  16. J. L. Doob. A Markov chain theorem. Probability and Statistics.(H. Cramer memorial volume) ed. U. Grenander. Almquist and Wiksel, Stockholm & New York, pages 50–57, 1959.
  17. J. L. Doob. Classical potential theory and its probabilistic counterpart, volume 262 of Grundlehren der Mathematischen Wissenschaften [Fundamental Principles of Mathematical Sciences]. Springer-Verlag, New York, 1984. ISBN 0-387-90881-1. doi: 10.1007/978-1-4612-5208-5.
  18. Traversing the Schrödinger bridge strait: Robert Fortet’s marvelous proof redux. Journal of Optimization Theory and Applications, 181(1):23–60, 2019.
  19. Wendell H Fleming. Exit probabilities and optimal stochastic control. Applied Mathematics and Optimization, 4:329–346, 1977.
  20. Wendell H Fleming. Logarithmic transformations and stochastic control. In Advances in Filtering and Optimal Stochastic Control: Proceedings of the IFIP-WG 7/1 Working Conference Cocoyoc, Mexico, February 1–6, 1982, pages 131–141. Springer, 2005.
  21. Deterministic and stochastic optimal control, volume 1. Springer Science & Business Media, 2012.
  22. Stochastic variational formula for fundamental solutions of parabolic pde. Applied Mathematics and Optimization, 13:193–204, 1985.
  23. H Föllmer. Random fields and diffusion processes. Ecole d’Ete de Probabilites de Saint-Flour XV-XVII, 1985-87, 1988.
  24. Robert Fortet. Résolution d’un système d’équations de m. Schrödinger. Journal de Mathématiques Pures et Appliquées, 19(1-4):83–105, 1940.
  25. Avner Friedman. Stochastic differential equations and applications. In Stochastic differential equations, pages 75–148. Springer, 1975.
  26. Positive contraction mappings for classical and quantum Schrödinger systems. Journal of Mathematical Physics, 56(3), 2015.
  27. Generative modeling for time series via Schrödinger bridge. arXiv preprint arXiv:2304.05093, 04 2023. doi: 10.13140/RG.2.2.25758.00324.
  28. Diffusion Schrödinger bridges for Bayesian computation. Statistical Science, 39(1):90–99, 2024.
  29. GANs trained by a two time-scale update rule converge to a local Nash equilibrium. Advances in Neural Information Processing Systems, 30, 2017.
  30. Denoising diffusion probabilistic models. Advances in Neural Information Processing Systems, 33:6840–6851, 2020.
  31. Schrödinger-Föllmer sampler: sampling without ergodicity. arXiv preprint arXiv:2106.10880, 2021.
  32. Aapo Hyvärinen. Estimation of non-normalized statistical models by score matching. Journal of Machine Learning Research, 6(4):695–709, 2005.
  33. Benton Jamison. The Markov processes of Schrödinger. Zeitschrift für Wahrscheinlichkeitstheorie und Verwandte Gebiete, 32(4):323–331, 1975.
  34. Brownian motion and stochastic calculus, volume 113. Springer Science & Business Media, 2012.
  35. Christian Léonard. A survey of the Schrödinger problem and some of its connections with optimal transport, 2013.
  36. Toshio Mikami. Variational processes from the weak forward equation. Communications in Mathematical Physics, 135:19–40, 1990.
  37. Fine-tuning diffusion models with limited data. In NeurIPS 2022 Workshop on Score-Based Methods, 2022.
  38. On free energy, stochastic control, and Schrödinger processes. In Modeling, Estimation and Control of Systems with Uncertainty: Proceedings of a Conference held in Sopron, Hungary, September 1990, pages 334–348. Springer, 1991.
  39. Stefano Peluchetti. Diffusion bridge mixture transports, Schrödinger bridge problems and generative modeling. arXiv preprint arXiv:2304.00917, 2023.
  40. On the Markov processes of Schrödinger, the Feynman-Kac formula and stochastic control. In Realization and Modelling in System Theory: Proceedings of the International Symposium MTNS-89, Volume I, pages 497–504. Springer, 1990.
  41. Improved sampling via learned diffusions. arXiv preprint arXiv:2307.01198, 2023.
  42. Improved techniques for training GANs. Advances in Neural Information Processing Systems, 29, 2016.
  43. Erwin Schrödinger. Über die Umkehrung der Naturgesetze. Verlag der Akademie der Wissenschaften in Kommission bei Walter De Gruyter u. Company, 1931.
  44. Erwin Schrödinger. Sur la théorie relativiste de l’électron et l’interprétation de la mécanique quantique. In Annales de l’institut Henri Poincaré, volume 2, pages 269–310, 1932.
  45. Conditional simulation using diffusion Schrödinger bridges. In Uncertainty in Artificial Intelligence, pages 1792–1802. PMLR, 2022.
  46. Schröwave: Realistic voice generation by solving two-stage conditional Schrödinger bridge problems. Digital Signal Processing, 141:104175, 2023.
  47. Ki-Ung Song. Applying regularized Schrödinger-bridge-based stochastic process in generative modeling. arXiv preprint arXiv:2208.07131, 2022.
  48. Generative modeling by estimating gradients of the data distribution. In Advances in Neural Information Processing Systems, pages 11895–11907, 2019.
  49. Score-based generative modeling through stochastic differential equations, 2021.
  50. Dual diffusion implicit bridges for image-to-image translation. arXiv preprint arXiv:2203.08382, 2022.
  51. Theoretical guarantees for sampling and inference in generative models with latent diffusions. In Conference on Learning Theory, pages 3084–3114. PMLR, 2019.
  52. Denoising diffusion samplers. In The Eleventh International Conference on Learning Representations, 2022.
  53. Pascal Vincent. A connection between score matching and denoising autoencoders. Neural Computation, 23(7):1661–1674, 2011.
  54. Deep generative learning via Schrödinger bridge. In Marina Meila and Tong Zhang, editors, Proceedings of the 38th International Conference on Machine Learning, volume 139 of Proceedings of Machine Learning Research, pages 10794–10804. PMLR, 18–24 Jul 2021.
  55. A score-based approach for training Schrödinger bridges for data modelling. Entropy, 25(2):316, 2023.
  56. Path integral sampler: A stochastic control approach for sampling. In International Conference on Learning Representations, 2021.

Summary

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

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