On the minimax optimality of Flow Matching through the connection to kernel density estimation
stat.ML, cs.LG, math.ST, stat.TH
Submitted: 2025-04-17
Updated: 2026-10-29
DOI: 10.1214/26-EJS2594
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Meta Flow Matching: Integrating Vector Fields on the Wasserstein Manifold
- Convergence of Diffusion Models Under the Manifold Hypothesis in High-Dimensions
- Gaussian-Channel Regularity and Distribution Learning with Flow Matching
- Theoretical Guarantees for High Order Trajectory Refinement in Generative Flows
- Distribution estimation via Flow Matching with Lipschitz guarantees
- Distribution learning via neural differential equations: a nonparametric statistical perspective
- On Flow Matching KL Divergence
Related papers
- Uniform-in-Time Weak Propagation-of-Chaos in Shallow Neural Networks
- Diffusion Models Observe Only Gradients: A Geometric Perspective on Score Matching Errors
- Quantifying Error Propagation and Model Collapse in Diffusion Models
- Behavior of prediction performance metrics with rare events
- Statistical Inference and Learning for Shapley Additive Explanations (SHAP)
- Control Variate Design for Score-Function Gradient Estimators in Importance-Weighted Variational Inference