tfp.mcmc: Modern Markov Chain Monte Carlo Tools Built for Modern Hardware

Abstract

Markov chain Monte Carlo (MCMC) is widely regarded as one of the most important algorithms of the 20th century. Its guarantees of asymptotic convergence, stability, and estimator-variance bounds using only unnormalized probability functions make it indispensable to probabilistic programming. In this paper, we introduce the TensorFlow Probability MCMC toolkit, and discuss some of the considerations that motivated its design.

arXiv commentBased on extended abstract submitted to PROBPROG 2020

References in corpus (4)

• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• MCMC using Hamiltonian dynamics
• NeuTra-lizing Bad Geometry in Hamiltonian Monte Carlo Using Neural Transport
• ELFI: Engine for Likelihood-Free Inference

Cited by in corpus (9)

• What Are Bayesian Neural Network Posteriors Really Like?
• Learning Functional Priors and Posteriors from Data and Physics
• A fast multi-fidelity method with uncertainty quantification for complex data correlations: Application to vortex-induced vibrations of marine risers
• How Good is the Bayes Posterior in Deep Neural Networks Really?
• Wavelet Flow: Fast Training of High Resolution Normalizing Flows
• Probabilistic Compute-in-Memory Design For Efficient Markov Chain Monte Carlo Sampling
• A Condition Number for Hamiltonian Monte Carlo
• Entropy-based adaptive Hamiltonian Monte Carlo
• FunMC: A functional API for building Markov Chains