Sampling from multi-modal distributions via stochastic localization

Add to your list(s) Download to your calendar using vCal

• Maxence Noble-Bourillot (Ecole Polytechnique Paris)
• Thursday 04 July 2024, 13:00-14:30
• External.

If you have a question about this talk, please contact nobody.

DML - Diffusions in machine learning: Foundations, generative models and non-convex optimisation

Building upon score-based learning, new interest in stochastic localization techniques has recently emerged. In these models, one seeks to noise a sample from the data distribution through a stochastic process, called observation process, and progressively learns a denoiser associated to this dynamics. Apart from specific applications, the use of stochastic localization for the problem of sampling from an unnormalized target density has not been explored extensively. This work contributes to fill this gap. We consider a general stochastic localization framework and introduce an explicit class of observation processes, associated with flexible denoising schedules. We provide a complete methodology, Stochastic Localization via Iterative Posterior Sampling (SLIPS), to obtain approximate samples of this dynamics, and as a by-product, samples from the target distribution. Our scheme is based on a Markov chain Monte Carlo estimation of the denoiser and comes with detailed practical guidelines. We highlight the bridge with recent methods built upon denoising models such as stochastic interpolants and diffusion models. Finally, we illustrate the benefits and applicability of SLIPS on several benchmarks of multi-modal distributions, including Gaussian mixtures in increasing dimensions, Bayesian logistic regression and a high-dimensional field system from statistical-mechanics. 

This talk is part of the Isaac Newton Institute Seminar Series series.

This talk is included in these lists:

• All CMS events
• External
• Featured lists
• INI info aggregator
• Isaac Newton Institute Seminar Series
• School of Physical Sciences
• bld31

Note that ex-directory lists are not shown.