中文
相关论文

相关论文: Interpolating between Optimal Transport and KL reg…

200 篇论文

Optimal Transport (OT) is being widely used in various fields such as machine learning and computer vision, as it is a powerful tool for measuring the similarity between probability distributions and histograms. In previous studies, OT has…

机器学习 · 统计学 2020-06-17 Yasunori Akagi , Yusuke Tanaka , Tomoharu Iwata , Takeshi Kurashima , Hiroyuki Toda

Regularized discrete optimal transport (OT) is a powerful tool to measure the distance between two discrete distributions that have been constructed from data samples on two different domains. While it has a wide range of applications in…

机器学习 · 计算机科学 2023-03-15 Yasutoshi Ida , Sekitoshi Kanai , Kazuki Adachi , Atsutoshi Kumagai , Yasuhiro Fujiwara

Divergences, also known as contrast functions, are distance-like quantities defined on manifolds of non-negative or probability measures. Using the duality in optimal transport, we introduce and study the one-parameter family of $L^{(\pm…

概率论 · 数学 2018-09-05 Ting-Kam Leonard Wong

In multi-task reinforcement learning, it is possible to improve the data efficiency of training agents by transferring knowledge from other different but related tasks. Because the experiences from different tasks are usually biased toward…

机器学习 · 计算机科学 2023-09-28 Bang Giang Le , Viet Cuong Ta

Optimal transport (OT) is a powerful framework to compare probability measures, a fundamental task in many statistical and machine learning problems. Substantial advances have been made in designing OT variants which are either…

机器学习 · 计算机科学 2025-02-04 Clément Bonet , Kimia Nadjahi , Thibault Séjourné , Kilian Fatras , Nicolas Courty

We propose a novel approach based on optimal transport (OT) for tackling the problem of highly mixed data in blind hyperspectral unmixing. Our method constrains the distribution of the estimated abundance matrix to resemble a targeted…

图像与视频处理 · 电气工程与系统科学 2025-09-26 D. Doutsas , B. Figliuzzi

We demonstrate the relevance of an algorithm called generalized iterative scaling (GIS) or simultaneous multiplicative algebraic reconstruction technique (SMART) and its rescaled block-iterative version (RBI-SMART) in the field of optimal…

数值分析 · 数学 2023-05-15 Johannes von Lindheim , Gabriele Steidl

We study the estimation of optimal transport (OT) maps between an arbitrary source probability measure and a log-concave target probability measure. Our contributions are twofold. First, we propose a new evolution equation in the set of…

最优化与控制 · 数学 2026-04-13 Théo Dumont , Théo Lacombe , François-Xavier Vialard

Optimal transport (OT) naturally arises in a wide range of machine learning applications but may often become the computational bottleneck. Recently, one line of works propose to solve OT approximately by searching the \emph{transport plan}…

机器学习 · 计算机科学 2021-11-15 Weijie Liu , Chao Zhang , Nenggan Zheng , Hui Qian

Minimization problems with respect to a one-parameter family of generalized relative entropies are studied. These relative entropies, which we term relative $\alpha$-entropies (denoted $\mathscr{I}_{\alpha}$), arise as redundancies under…

信息论 · 计算机科学 2015-06-11 M. Ashok Kumar , Rajesh Sundaresan

We provide a computational complexity analysis for the Sinkhorn algorithm that solves the entropic regularized Unbalanced Optimal Transport (UOT) problem between two measures of possibly different masses with at most $n$ components. We show…

计算复杂性 · 计算机科学 2020-11-20 Khiem Pham , Khang Le , Nhat Ho , Tung Pham , Hung Bui

The forward Kullback-Leibler (KL) divergence is a ubiquitous objective for fitting a parameterized distribution to samples due to its tractability and equivalence to maximum likelihood estimation (MLE). Its inherent asymmetry, however, may…

机器学习 · 计算机科学 2026-05-12 Omri Ben-Dov , Luiz F. O. Chamon

In this paper, we introduce a neural network-based method to address the high-dimensional dynamic unbalanced optimal transport (UOT) problem. Dynamic UOT focuses on the optimal transportation between two densities with unequal total mass,…

最优化与控制 · 数学 2024-09-23 Wei Wan , Jiangong Pan , Yuejin Zhang , Chenglong Bao , Zuoqiang Shi

We propose a new regularized optimal transport (OT) formulation, termed sliced-regularized optimal transport (SROT). Unlike entropic OT (EOT), which regularizes the transport plan toward an independent coupling, SROT regularizes it toward a…

机器学习 · 统计学 2026-05-21 Khai Nguyen

Several recent applications of optimal transport (OT) theory to machine learning have relied on regularization, notably entropy and the Sinkhorn algorithm. Because matrix-vector products are pervasive in the Sinkhorn algorithm, several…

机器学习 · 统计学 2021-03-09 Meyer Scetbon , Marco Cuturi , Gabriel Peyré

This paper presents a novel two-step approach for the fundamental problem of learning an optimal map from one distribution to another. First, we learn an optimal transport (OT) plan, which can be thought as a one-to-many map between the two…

We introduce a new class of convex-regularized Optimal Transport losses, which generalizes the classical Entropy-regularization of Optimal Transport and Sinkhorn divergences, and propose a generalized Sinkhorn algorithm. Our framework…

最优化与控制 · 数学 2020-07-03 Simone Di Marino , Augusto Gerolin

An efficient method for computing solutions to the Optimal Transportation (OT) problem with a wide class of cost functions is presented. The standard linear programming (LP) discretization of the continuous problem becomes intractible for…

数值分析 · 数学 2015-09-15 Adam M. Oberman , Yuanlong Ruan

Variational problems that involve Wasserstein distances and more generally optimal transport (OT) theory are playing an increasingly important role in data sciences. Such problems can be used to form an examplar measure out of various…

机器学习 · 计算机科学 2018-11-15 Marco Cuturi , Gabriel Peyré

We give an overview of various results and methods related to information-theoretic distances of R\'enyi type in the light of their applications to the central limit theorem (CLT). The first part (Sections 1-9) is devoted to the total…

信息论 · 计算机科学 2025-03-07 S. G. Bobkov , G. P. Chistyakov , F. Götze