Variational Online Mirror Descent for Robust Learning in Schrödinger Bridge

• URL: http://arxiv.org/abs/2504.02618v2
• Date: Tue, 08 Apr 2025 17:49:16 GMT
• Title: Variational Online Mirror Descent for Robust Learning in Schrödinger Bridge
• Authors: Dong-Sig Han, Jaein Kim, Hee Bin Yoo, Byoung-Tak Zhang,
• Abstract summary: Sch"odinger bridge (SB) has evolved into a universal class of probabilistic generative models.<n>Recent studies regarding the Sinkhorn algorithm through mirror descent (MD) have gained attention.<n>We propose a variational online MD (OMD) framework for the SB problems, which provides further stability to SB solvers.
• Score: 22.516011429301017
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Sch\"odinger bridge (SB) has evolved into a universal class of probabilistic generative models. In practice, however, estimated learning signals are often uncertain, and the reliability promised by existing methods is often based on speculative optimal-case scenarios. Recent studies regarding the Sinkhorn algorithm through mirror descent (MD) have gained attention, revealing geometric insights into solution acquisition of the SB problems. In this paper, we propose a variational online MD (OMD) framework for the SB problems, which provides further stability to SB solvers. We formally prove convergence and a regret bound for the novel OMD formulation of SB acquisition. As a result, we propose a simulation-free SB algorithm called Variational Mirrored Schr\"odinger Bridge (VMSB) by utilizing the Wasserstein-Fisher-Rao geometry of the Gaussian mixture parameterization for Schr\"odinger potentials. Based on the Wasserstein gradient flow theory, the algorithm offers tractable learning dynamics that precisely approximate each OMD step. In experiments, we validate the performance of the proposed VMSB algorithm across an extensive suite of benchmarks. VMSB consistently outperforms contemporary SB solvers on a range of SB problems, demonstrating the robustness predicted by our theory.

Related papers

• Latent Schrodinger Bridge: Prompting Latent Diffusion for Fast Unpaired Image-to-Image Translation [58.19676004192321]
  Diffusion models (DMs), which enable both image generation from noise and inversion from data, have inspired powerful unpaired image-to-image (I2I) translation algorithms. We tackle this problem with Schrodinger Bridges (SBs), which are differential equations (SDEs) between distributions with minimal transport cost. Inspired by this observation, we propose Latent Schrodinger Bridges (LSBs) that approximate the SB ODE via pre-trained Stable Diffusion. We demonstrate that our algorithm successfully conduct competitive I2I translation in unsupervised setting with only a fraction of cost required by previous DM-
  arXiv  Detail & Related papers  (2024-11-22T11:24:14Z)
• Light and Optimal Schrödinger Bridge Matching [67.93806073192938]
  We propose a novel procedure to learn Schr"odinger Bridges (SB) which we call the textbf Schr"odinger bridge matching. We show that the optimal bridge matching objective coincides with the recently discovered energy-based modeling (EBM) objectives to learn EOT/SB. We develop a light solver (which we call LightSB-M) to implement optimal matching in practice using the mixture parameterization of the Schr"odinger potential.
  arXiv  Detail & Related papers  (2024-02-05T17:17:57Z)
• Light Schrödinger Bridge [72.88707358656869]
  We develop a lightweight, simulation-free and theoretically justified Schr"odinger Bridges solver. Our light solver resembles the Gaussian mixture model which is widely used for density estimation. Inspired by this similarity, we also prove an important theoretical result showing that our light solver is a universal approximator of SBs.
  arXiv  Detail & Related papers  (2023-10-02T13:06:45Z)
• Building the Bridge of Schr\"odinger: A Continuous Entropic Optimal Transport Benchmark [96.06787302688595]
  We propose a novel way to create pairs of probability distributions for which the ground truth OT solution is known by the construction. We use these benchmark pairs to test how well existing neural EOT/SB solvers actually compute the EOT solution.
  arXiv  Detail & Related papers  (2023-06-16T20:03:36Z)
• Unbalanced Diffusion Schr\"odinger Bridge [71.31485908125435]
  We introduce unbalanced DSBs which model the temporal evolution of marginals with arbitrary finite mass. This is achieved by deriving the time reversal of differential equations with killing and birth terms. We present two novel algorithmic schemes that comprise a scalable objective function for training unbalanced DSBs.
  arXiv  Detail & Related papers  (2023-06-15T12:51:56Z)
• Diffusion Schr\"odinger Bridge Matching [36.95088080680221]
  We introduce Iterative Markovian Fitting (IMF) and Diffusion Schr"odinger Bridge Matching (DSBM) IMF is a new methodology for solving SB problems, and DSBM is a novel numerical algorithm for computing IMF iterates. We demonstrate the performance of DSBM on a variety of problems.
  arXiv  Detail & Related papers  (2023-03-29T16:59:22Z)
• The Schr\"odinger Bridge between Gaussian Measures has a Closed Form [101.79851806388699]
  We focus on the dynamic formulation of OT, also known as the Schr"odinger bridge (SB) problem. In this paper, we provide closed-form expressions for SBs between Gaussian measures.
  arXiv  Detail & Related papers  (2022-02-11T15:59:01Z)
• Likelihood Training of Schr\"odinger Bridge using Forward-Backward SDEs Theory [29.82841891919951]
  It remains unclear whether the optimization principle of SB relates to the modern training of deep generative models. We present a novel computational framework for likelihood training of SB models grounded on Forward-Backward Theory. We show that the resulting training achieves comparable results on generating realistic images on MNIST, CelebA, and CIFAR10.
  arXiv  Detail & Related papers  (2021-10-21T17:18:59Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.