Accelerating MCMC by Rare Intermittent Resets

Vivek S. Borkar; Syomantak Chaudhuri

Performance Evaluation Methodologies and Tools. 14th EAI International Conference, VALUETOOLS 2021, Virtual Event, October 30–31, 2021, Proceedings

Research Article

Accelerating MCMC by Rare Intermittent Resets

Download

4 downloads

Cite: BibTeX Plain Text

@INPROCEEDINGS{10.1007/978-3-030-92511-6_7,
    author={Vivek S. Borkar and Syomantak Chaudhuri},
    title={Accelerating MCMC by Rare Intermittent Resets},
    proceedings={Performance Evaluation Methodologies and Tools. 14th EAI International Conference, VALUETOOLS 2021, Virtual Event, October 30--31, 2021, Proceedings},
    proceedings_a={VALUETOOLS},
    year={2021},
    month={12},
    keywords={Markov Chain Monte Carlo Time inhomogeneous Markov chain Rare resets Accelerated convergence Martingale law of large numbers Concentration bounds},
    doi={10.1007/978-3-030-92511-6_7}
}

Vivek S. Borkar
Syomantak Chaudhuri
Year: 2021
Accelerating MCMC by Rare Intermittent Resets
VALUETOOLS
Springer
DOI: 10.1007/978-3-030-92511-6_7

Vivek S. Borkar¹^,*, Syomantak Chaudhuri¹

1: Department of Electrical Engineering

*Contact email: borkar@iitb.ac.in

Abstract

We propose a scheme for accelerating Markov Chain Monte Carlo by introducing random resets that become increasingly rare in a precise sense. We show that this still leads to the desired asymptotic average and establish an associated concentration bound. We show by numerical experiments that this scheme can be used to advantage in order to accelerate convergence by a judicious choice of the resetting mechanism.

Keywords: Markov Chain Monte Carlo Time inhomogeneous Markov chain Rare resets Accelerated convergence Martingale law of large numbers Concentration bounds

Published: 2021-12-08
Appears in: SpringerLink

: http://dx.doi.org/10.1007/978-3-030-92511-6_7

Accelerating MCMC by Rare Intermittent Resets

Abstract

About EAI

Community

Publish with EAI