tfp.mcmc: Modern Markov Chain Monte Carlo Tools Built for Modern Hardware
Computation
2020-02-05 v1 Programming Languages
Machine Learning
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.
Keywords
Cite
@article{arxiv.2002.01184,
title = {tfp.mcmc: Modern Markov Chain Monte Carlo Tools Built for Modern Hardware},
author = {Junpeng Lao and Christopher Suter and Ian Langmore and Cyril Chimisov and Ashish Saxena and Pavel Sountsov and Dave Moore and Rif A. Saurous and Matthew D. Hoffman and Joshua V. Dillon},
journal= {arXiv preprint arXiv:2002.01184},
year = {2020}
}
Comments
Based on extended abstract submitted to PROBPROG 2020