Keywords: stochastic gradient Langevin dynamics, MCMC, importance sampling, Wang-Landau algorithm, Parallel MCMC Methods, stochastic approximation
Abstract: We propose an interacting contour stochastic gradient Langevin dynamics (ICSGLD) sampler, an embarrassingly parallel multiple-chain contour stochastic gradient Langevin dynamics (CSGLD) sampler with efficient interactions. We show that ICSGLD can be theoretically more efficient than a single-chain CSGLD with an equivalent computational budget. We also present a novel random-field function, which facilitates the estimation of self-adapting parameters in big data and obtains free mode explorations. Empirically, we compare the proposed algorithm with popular benchmark methods for posterior sampling. The numerical results show a great potential of ICSGLD for large-scale uncertainty estimation tasks.
One-sentence Summary: We propose an interacting contour stochastic gradient Langevin dynamics sampler and prove it can be theoretically more efficient than a single-chain process with an equivalent computational budget.
Supplementary Material: zip
Community Implementations: [![CatalyzeX](/images/catalyzex_icon.svg) 1 code implementation](https://www.catalyzex.com/paper/arxiv:2202.09867/code)