paper-with-me

홈 › Papers

Understanding MCMC Dynamics as Flows on the Wasserstein Space

2019-02-01 · Chang Liu, Jingwei Zhuo, Jun Zhu

It is known that the Langevin dynamics used in MCMC is the gradient flow of the KL divergence on the Wasserstein space, which helps convergence analysis and inspires recent particle-based variational inference methods (ParVIs). But no more MCMC dynamics is understood in this way. In this work, by developing novel concepts, we propose a theoretical framework that recognizes a general MCMC dynamics as the fiber-gradient Hamiltonian flow on the Wasserstein space of a fiber-Riemannian Poisson manifold. The "conservation + convergence" structure of the flow gives a clear picture on the behavior of general MCMC dynamics. The framework also enables ParVI simulation of MCMC dynamics, which enriches the ParVI family with more efficient dynamics, and also adapts ParVI advantages to MCMCs. We develop two ParVI methods for a particular MCMC dynamics and demonstrate the benefits in experiments.

📄 PDF Abstract BibTeX arXiv:1902.00282

Code (1)

chang-ml-thu/FGH-flow 공식 구현 tf

Tasks

Novel ConceptsVariational Inference

Similar Papers 제목 키워드 기반

Variational inference via Wasserstein gradient flows

2022-05-31 · Marc Lambert, Sinho Chewi, Francis Bach, Silvère Bonnabel 외

Along with Markov chain Monte Carlo (MCMC) methods, variational inference (VI) has emerged as a central computational approach to large-scale Bayesian inference. Rather than sampling from the true posterior $\pi$, VI aim…

Bayesian InferenceVariational Inference

De-randomizing MCMC dynamics with the diffusion Stein operator

2021-10-07 · NeurIPS 2021 12 · Zheyang Shen, Markus Heinonen, Samuel Kaski

Approximate Bayesian inference estimates descriptors of an intractable target distribution - in essence, an optimization problem within a family of distributions. For example, Langevin dynamics (LD) extracts asymptotical…

Bayesian Inference

Couplings for Andersen Dynamics

2020-09-29 · Nawaf Bou-Rabee, Andreas Eberle

Andersen dynamics is a standard method for molecular simulations, and a precursor of the Hamiltonian Monte Carlo algorithm used in MCMC inference. The stochastic process corresponding to Andersen dynamics is a PDMP (piec…

Accelerated Information Gradient flow

2019-09-04 · Yifei Wang, Wuchen Li

We present a framework for Nesterov's accelerated gradient flows in probability space to design efficient mean-field Markov chain Monte Carlo (MCMC) algorithms for Bayesian inverse problems. Here four examples of informa…

Bayesian Inference

Information Newton's flow: second-order optimization method in probability space

2020-01-13 · Yifei Wang, Wuchen Li

We introduce a framework for Newton's flows in probability space with information metrics, named information Newton's flows. Here two information metrics are considered, including both the Fisher-Rao metric and the Wasse…