9 papers
Radial Tail Repair: Post-Training Correction for Diffusion and Flow Models
Hamza Cherkaoui, Hélène Halconruy, Antonio Ocello
Diffusion and flow-matching models may struggle to accurately represent distributional tails, especially for heavy-tailed data. Existing approaches often use heavy-tailed noise or…
Theoretical Guarantees for SMC-Guided Diffusion Sampling
Stanislas Strasman, Gabriel Victorino Cardoso, Sylvain Le Corff +2
Post-hoc conditioning of pretrained diffusion models can be addressed using Sequential Monte Carlo (SMC) methods. By evolving an interacting particle system, SMC-guided diffusion s…
Do Heavy Tails Help Diffusion? On the Subtle Trade-off Between Initialization and Training
Hamza Cherkaoui, Hélène Halconruy, Antonio Ocello
Recent works have proposed incorporating heavy-tailed (HT) noise into diffusion- and flow-based generative models, with the goals of better recovering the tails of target distribut…
On Forgetting and Stability of Score-based Generative models
Stanislas Strasman, Gabriel Cardoso, Sylvain Le Corff +2
Understanding the stability and long-time behavior of generative models is a fundamental problem in modern machine learning. This paper provides quantitative bounds on the sampling…
Wasserstein Convergence of Critically Damped Langevin Diffusions
Stanislas Strasman, Sobihan Surendran, Claire Boyer +3
Score-based Generative Models (SGMs) have achieved impressive performance in data generation across a wide range of applications and benefit from strong theoretical guarantees. Rec…
Finite-Sample Convergence Bounds for Trust Region Policy Optimization in Mean-Field Games
Antonio Ocello, Daniil Tiapkin, Lorenzo Mancini +2
We introduce Mean-Field Trust Region Policy Optimization (MF-TRPO), a novel algorithm designed to compute approximate Nash equilibria for ergodic Mean-Field Games (MFG) in finite s…
Bit-Level Discrete Diffusion with Markov Probabilistic Models: An Improved Framework with Sharp Convergence Bounds under Minimal Assumptions
Le-Tuyet-Nhi Pham, Dario Shariatian, Antonio Ocello +2
This paper introduces Discrete Markov Probabilistic Models (DMPMs), a novel discrete diffusion algorithm for discrete data generation. The algorithm operates in discrete bit space,…
Beyond Log-Concavity and Score Regularity: Improved Convergence Bounds for Score-Based Generative Models in W2-distance
Marta Gentiloni-Silveri, Antonio Ocello
Score-based Generative Models (SGMs) aim to sample from a target distribution by learning score functions using samples perturbed by Gaussian noise. Existing convergence bounds for…
An analysis of the noise schedule for score-based generative models
Stanislas Strasman, Antonio Ocello, Claire Boyer +2
Score-based generative models (SGMs) aim at estimating a target data distribution by learning score functions using only noise-perturbed samples from the target.Recent literature h…