activity
20232026
collaborators
Showing stat.MLShow all

7 papers · 1 filter

stat.ML2026

PAC-Bayesian Reconstruction Guarantees for Time Series Variational Autoencoders

Chloé Hashimoto-Cullen, Ghislain Agoua, Benjamin Guedj +1

Forecasting time series accurately is critical for applications with complex data ranging from energy systems to healthcare and finance. Among current state of the art models, gene…

stat.ML2026

Non-Asymptotic Error Bounds for SMC with Biased Proposals: Application to Conditional Diffusion Sampling

Stanislas Strasman, Gabriel Victorino Cardoso, Sylvain Le Corff +2

Sequential Monte Carlo (SMC) methods are a natural tool for post-hoc conditioning of pretrained generative models, but in many applications the mutation kernels used by the particl…

stat.ML2026

Non-asymptotic Convergence of Stochastic Gradient Descent in Score-based Generative Models

Stanislas Strasman, Sobihan Surendran, Sylvain Le Corff

Score-based Generative Models (SGMs) have achieved impressive performance in data generation across a wide range of applications. While the statistical properties of their sampling…

stat.ML2026

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…

stat.ML2025

Latent Guided Sampling for Combinatorial Optimization

Sobihan Surendran, Adeline Fermanian, Sylvain Le Corff

Combinatorial Optimization problems are widespread in domains such as logistics, manufacturing, and drug discovery, yet their NP-hard nature makes them computationally challenging.…

stat.ML2024

Diffusion posterior sampling for simulation-based inference in tall data settings

Julia Linhart, Gabriel Victorino Cardoso, Alexandre Gramfort +2

Identifying the parameters of a non-linear model that best explain observed data is a core task across scientific fields. When such models rely on complex simulators, evaluating th…