Proximal Interacting Particle Langevin Algorithms

arXiv:2406.14292v1 Announce Type: cross
Abstract: We introduce a class of algorithms, termed Proximal Interacting Particle Langevin Algorithms (PIPLA), for inference and learning in latent variable models whose joint probability density is non-differentiable. Leveraging proximal Markov chain Monte Carlo (MCMC) techniques and the recently introduced interacting particle Langevin algorithm (IPLA), we propose several variants within the novel proximal IPLA family, tailored to the problem of estimating parameters in a non-differentiable statistical model. We prove nonasymptotic bounds for the parameter estimates produced …

abstract algorithm algorithms arxiv class differentiable inference markov math.oc mcmc particle probability stat.co stat.ml type variants