Neural Optimal Transport with Lagrangian Costs

• URL: http://arxiv.org/abs/2406.00288v1
• Date: Sat, 1 Jun 2024 03:34:00 GMT
• Title: Neural Optimal Transport with Lagrangian Costs
• Authors: Aram-Alexandre Pooladian, Carles Domingo-Enrich, Ricky T. Q. Chen, Brandon Amos,
• Abstract summary: We investigate the optimal transport problem between probability measures when the underlying cost function is understood to satisfy a Lagrangian cost. Our contributions are of computational interest, where we demonstrate the ability to efficiently compute geodesics and amortize spline-based paths. Unlike prior work, we also output the resulting Lagrangian optimal transport map without requiring an ODE solver.
• Score: 29.091068250865504
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We investigate the optimal transport problem between probability measures when the underlying cost function is understood to satisfy a least action principle, also known as a Lagrangian cost. These generalizations are useful when connecting observations from a physical system where the transport dynamics are influenced by the geometry of the system, such as obstacles (e.g., incorporating barrier functions in the Lagrangian), and allows practitioners to incorporate a priori knowledge of the underlying system such as non-Euclidean geometries (e.g., paths must be circular). Our contributions are of computational interest, where we demonstrate the ability to efficiently compute geodesics and amortize spline-based paths, which has not been done before, even in low dimensional problems. Unlike prior work, we also output the resulting Lagrangian optimal transport map without requiring an ODE solver. We demonstrate the effectiveness of our formulation on low-dimensional examples taken from prior work. The source code to reproduce our experiments is available at https://github.com/facebookresearch/lagrangian-ot.

Related papers

• OTClean: Data Cleaning for Conditional Independence Violations using Optimal Transport [51.6416022358349]
  sys is a framework that harnesses optimal transport theory for data repair under Conditional Independence (CI) constraints. We develop an iterative algorithm inspired by Sinkhorn's matrix scaling algorithm, which efficiently addresses high-dimensional and large-scale data.
  arXiv  Detail & Related papers  (2024-03-04T18:23:55Z)
• Conditional Optimal Transport on Function Spaces [53.9025059364831]
  We develop a theory of constrained optimal transport problems that describe block-triangular Monge maps. This generalizes the theory of optimal triangular transport to separable infinite-dimensional function spaces with general cost functions. We present numerical experiments that demonstrate the computational applicability of our theoretical results for amortized and likelihood-free inference of functional parameters.
  arXiv  Detail & Related papers  (2023-11-09T18:44:42Z)
• A Computational Framework for Solving Wasserstein Lagrangian Flows [48.87656245464521]
  In general, the optimal density path is unknown, and solving these variational problems can be computationally challenging. We propose a novel deep learning based framework approaching all of these problems from a unified perspective. We showcase the versatility of the proposed framework by outperforming previous approaches for the single-cell trajectory inference.
  arXiv  Detail & Related papers  (2023-10-16T17:59:54Z)
• Large-Scale OD Matrix Estimation with A Deep Learning Method [70.78575952309023]
  The proposed method integrates deep learning and numerical optimization algorithms to infer matrix structure and guide numerical optimization. We conducted tests to demonstrate the good generalization performance of our method on a large-scale synthetic dataset.
  arXiv  Detail & Related papers  (2023-10-09T14:30:06Z)
• Entropic Neural Optimal Transport via Diffusion Processes [105.34822201378763]
  We propose a novel neural algorithm for the fundamental problem of computing the entropic optimal transport (EOT) plan between continuous probability distributions. Our algorithm is based on the saddle point reformulation of the dynamic version of EOT which is known as the Schr"odinger Bridge problem. In contrast to the prior methods for large-scale EOT, our algorithm is end-to-end and consists of a single learning step.
  arXiv  Detail & Related papers  (2022-11-02T14:35:13Z)
• Constrained Mass Optimal Transport [0.0]
  This paper introduces the problem of constrained optimal transport. A family of algorithms is introduced to solve a class of constrained saddle point problems. Convergence proofs and numerical results are presented.
  arXiv  Detail & Related papers  (2022-06-05T06:47:25Z)
• Online Learning to Transport via the Minimal Selection Principle [2.3857747529378917]
  We study the Online Learning Transport (OLT) problem where the decision variable is a convex, an-dimensional object. We derive a novel method called the minimal selection or exploration (SoMLT) algorithm to solve OLT problems using mean-field and discretization techniques.
  arXiv  Detail & Related papers  (2022-02-09T21:25:58Z)
• Do Neural Optimal Transport Solvers Work? A Continuous Wasserstein-2 Benchmark [133.46066694893318]
  We evaluate the performance of neural network-based solvers for optimal transport. We find that existing solvers do not recover optimal transport maps even though they perform well in downstream tasks.
  arXiv  Detail & Related papers  (2021-06-03T15:59:28Z)
• Functional optimal transport: map estimation and domain adaptation for functional data [35.60475201744369]
  We introduce a formulation of optimal transport problem for distributions on function spaces. For numerous machine learning tasks, data can be naturally viewed as samples drawn from spaces of functions. We develop an efficient algorithm for finding the transport map between functional domains.
  arXiv  Detail & Related papers  (2021-02-07T19:29:28Z)

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.