cs.LGNov 14, 2026

Explicit Regularization for Flow Matching

Authors: Lars Ruthotto, Eran Treister

Organizations: Departments of Mathematics and Computer Science, Emory University, Atlanta, GA, USA · Faculty of Computer and Information Science, Ben-Gurion University of the Negev, Beer-Sheva, Israel

Abstract

Learning the underlying probability distribution from samples is a core problem in statistics and applied mathematics. Generative models based on diffusion and flow matching learn a velocity field whose flow transports a known distribution to the data distribution. They produce novel, realistic samples, although the empirical learning problem is ill posed near the data distribution. Its minimizer transports the samples onto the training data, and small perturbations of the data change the velocity field strongly near the end of the flow. The observed generalization is often attributed to implicit regularization by the network parameterization, early stopping of the training, and the time integration during sampling. Explicit regularization has advanced the theory and computation of many inverse problems and has recently been proposed to prevent memorization in generative models. With a regularizer on the magnitude of the velocity, however, the field approaches the memorizing velocity unless the regularization parameter grows to infinity fast enough, and it then stays bounded by a constant that is small only for large parameters. We show that penalizing the derivatives of order kk of the velocity renders the DD-dimensional empirical problem well posed uniformly in time if 2k>D2k > D and the training points are in general position. Our experiments in one and two dimensions discretize the velocity with splines and use direct solvers for the regularized least-squares problems to exclude optimization error. Refining the discretization in space and time changes the error of the Hessian-regularized flows only marginally, whereas without the regularizer the error tends to that of memorization. These flows generalize, with errors comparable to or lower than those of well-tuned kernel density estimates.

Figures & tables

Explore similar work

Nov 1, 2026

Generative Modeling with Continuous Flows: Sample Complexity of Flow Matching

Flow matching has emerged as a promising alternative to diffusion-based generative modeling, offering a continuous-time ordinary differential equation formulation with fast sampling and a simple regression-based training objective. Despite its empirical success, some theoretical analyses idealize the learning procedure by assuming that the velocity field is obtained by exact empirical risk minimization. This assumption does not reflect the practical finite-dataset setting, where the empirical flow-matching loss is only approximately minimized by a finite number of stochastic-gradient updates. We develop an empirical-risk-minimizer-free analysis for flow matching that explicitly accounts for both finite-sample statistical error and finite-time empirical optimization error. Our analysis considers a finite dataset of NN flow-matching samples and a learned parameter θTθ_T obtained after TT stochastic gradient updates on the empirical objective. Under regularity assumptions on the interpolation path, sub-exponential source and target distributions, approximation by a bounded-parameter smooth neural velocity class, empirical-loss optimization, and uniform flow stability, we prove a guarantee that holds with high probability over the finite training dataset and in expectation over the stochastic-gradient randomness. The resulting bound separates approximation error, statistical error from the finite dataset, tail error from unbounded source and target distributions, and empirical optimization error from approximate minimization of the finite-sample objective.
Oct 2, 2026

Generalization Bounds for Flow-matching Generative Models for Intrinsically Low-dimensional Data

Despite the remarkable empirical success of flow-matching models, their statistical generalization guarantees remain underdeveloped. Existing analyses often impose restrictive assumptions on the estimated velocity field and yield convergence rates that fail to reflect the intrinsic low-dimensional structure common in real data, such as natural images and molecular geometries. In this work, we study the statistical generalization of flow-matching models for learning an unknown distribution PdataP_{\mathrm{data}} from finitely many samples. We derive finite-sample error bounds on the learned generative distribution, measured in the Wasserstein-pp distance, for all p≥1p\geq 1. Specifically, given nn i.i.d. samples from PdataP_{\mathrm{data}}, we show that, for every d>dp∗(Pdata)d>d_p^\ast(P_{\mathrm{data}}) and appropriately chosen network architectures and hyperparameters, the learned distribution P^FM\widehat{P}^{\mathrm{FM}} satisfies Wp(P^FM,Pdata)≲n−1/d+n−1/(2p)(log⁡(1/ξ))1/(2p) \mathbb{W}_p(\widehat{P}^{\mathrm{FM}},P_{\mathrm{data}}) \lesssim n^{-1/d}+n^{-1/(2p)}\bigl(\log(1/ξ)\bigr)^{1/(2p)} with probability at least 1−ξ1-ξ, where dp∗(Pdata)d_p^\ast(P_{\mathrm{data}}) denotes the Wasserstein-pp dimension of the target measure. Our results demonstrate that flow matching naturally adapts to the intrinsic geometry of data and mitigates the curse of dimensionality, as the convergence exponent depends on the intrinsic rather than ambient dimension. These guarantees remain meaningful in high-dimensional regimes and provide a theoretical explanation for the empirical success of flow matching on structured data distributions under substantially more relaxed assumptions than those in existing analyses.
Sep 30, 2026

Flow Matching under Noisy Latent Structure: Beyond Exact Low-Dimensional Support

Flow Matching (FM) learns a velocity field whose ODE transports a simple source distribution to a target law. Existing finite-sample theory largely treats ambient-space regularity or data supported exactly on low-dimensional sets. We study linear FM under a noisy latent-generator model, where a low-dimensional Hölder map is perturbed by nondegenerate ambient Gaussian noise, so the target law is full-dimensional despite its latent structure. We construct a spatially regular ReLU velocity class and establish non-asymptotic high-probability approximation and estimation bounds whose leading sample-size exponent is governed by the latent dimension rather than the ambient dimension, with ambient and noise dependence kept explicit. Fixed positive target noise keeps the interpolation nondegenerate over the full time interval. The same spatial regularity propagates the learned velocity error through the transport ODE, yielding a corresponding Wasserstein convergence guarantee. These results show that exact low-dimensional support is not necessary for Flow Matching to retain latent-dimensional statistical behavior.