arXiv Machine Learning

Generative Modeling by Minimizing the Wasserstein-2 Loss

This paper introduces a generative model that minimizes the second‑order Wasserstein loss (W₂) by solving a distribution‑dependent ordinary differential equation (ODE) whose dynamics involve the Kantorovich potential of the true data distribution and its current estimate. The authors prove that the time‑marginal laws of this ODE form a gradient flow for the W₂ loss, converging exponentially to the true data distribution, and propose an Euler scheme that recovers this gradient flow in the limit. An algorithm based on this scheme, combined with persistent training, is shown in experiments to outperform Wasserstein GANs in both low‑ and high‑dimensional settings when the level of persistent training is appropriately increased.

arXiv Machine Learning
Oct 19

Deep conditional distribution learning via conditional F\"ollmer flow

The paper presents Conditional Föllmer Flow, an ODE‑based deep generative approach for learning conditional distributions. Starting from a Gaussian, the flow approximates the target distribution near time 1, with Euler discretization and a neural‑network‑estimated velocity field. The authors prove convergence in Wasserstein‑2 distance and demonstrate superior performance on both nonparametric density estimation and image‑based tasks compared to existing methods.

By Jinyuan Chang, Zhao Ding, Yuling Jiao, Ruoxuan Li, Jerry Zhijian Yang
arXiv Statistics ML
Oct 2

Convergence Analysis of the Wasserstein Proximal Algorithm beyond Geodesic Convexity

The paper presents a convergence analysis of the Wasserstein proximal algorithm without requiring geodesic convexity of the objective functional. By assuming a Wasserstein analog of the Euclidean Polyak–Łojasiewicz inequality, the authors prove that the algorithm achieves unbiased, linear convergence, improving on previous rates that required strong geodesic convexity. The analysis is extended to inexact proximal steps for geodesically semiconvex objectives, and numerical experiments show faster convergence than noisy gradient descent on mean‑field neural networks.

By Shuailong Zhu, Xiaohui Chen
arXiv Statistics ML
Oct 16

Optimal score function estimation via derivatives constraints

The paper develops a statistical framework showing that empirical score matching can achieve minimax optimal performance for diffusion generative models when data lie on a low‑dimensional manifold. It establishes convergence rates for empirical risk minimization over constrained Sobolev classes and demonstrates that the resulting score estimators yield minimax optimal Wasserstein‑1 rates for the generated distribution, up to logarithmic factors. The analysis reveals that the statistical rates depend on the intrinsic dimension of the data manifold rather than its ambient dimension, providing conditions under which score‑based generative modeling generalizes rather than overfits.

By Thomas Bonis, Thanh Mai Pham Ngoc, Viet Chi Tran
arXiv Machine Learning
Oct 9

Barron Optimal Transport I: Generative Modeling

The paper introduces a new optimal transport framework that uses the Barron energy—an intrinsic neural network complexity measure—instead of the traditional kinetic energy. By modeling transport as deep residual networks that map a reference distribution to a target distribution, the authors define a new metric on probability measures and analyze its properties for generative modeling. They demonstrate that diffusion generative models are suboptimal in this Barron geometry and explore how adaptivity can improve alternative generative approaches.

By Evan Dogariu, Joan Bruna