Schr{ö}dinger-F{ö}llmer Sampler: Sampling without Ergodicity
arXiv:2106.10880
Abstract
Sampling from probability distributions is an important problem in statistics and machine learning, specially in Bayesian inference when integration with respect to posterior distribution is intractable and sampling from the posterior is the only viable option for inference. In this paper, we propose Schrödinger-Föllmer sampler (SFS), a novel approach for sampling from possibly unnormalized distributions. The proposed SFS is based on the Schrödinger-Föllmer diffusion process on the unit interval with a time dependent drift term, which transports the degenerate distribution at time zero to the target distribution at time one. Comparing with the existing Markov chain Monte Carlo samplers that require ergodicity, no such requirement is needed for SFS. Computationally, SFS can be easily implemented using the Euler-Maruyama discretization. In theoretical analysis, we establish non-asymptotic error bounds for the sampling distribution of SFS in the Wasserstein distance under suitable conditions. We conduct numerical experiments to evaluate the performance of SFS and demonstrate that it is able to generate samples with better quality than several existing methods.
References in corpus (16)
- MCMC using Hamiltonian dynamics
- A Conceptual Introduction to Hamiltonian Monte Carlo
- Stochastic Gradient Hamiltonian Monte Carlo
- Rejection-free Monte-Carlo sampling for general potentials
- Bayesian Posterior Sampling via Stochastic Gradient Fisher Scoring
- Convergence of Langevin MCMC in KL-divergence
- Probabilistic Inference on Noisy Time Series (PINTS)
- Poincaré and logarithmic Sobolev inequalities by decomposition of the energy landscape
- On stochastic gradient Langevin dynamics with dependent data streams: the fully non-convex case
- Does Hamiltonian Monte Carlo mix faster than a random walk on multimodal densities?
- Nonasymptotic estimates for Stochastic Gradient Langevin Dynamics under local conditions in nonconvex optimization
- Log-Sobolev inequalities: Different roles of Ric and Hess
- The data-driven Schroedinger bridge
- Schrödinger Bridge Samplers
- Stochastic control liaisons: Richard Sinkhorn meets Gaspard Monge on a Schroedinger bridge
- Deep Generative Learning via Schrödinger Bridge