Can I read Adaptive Parallel Tempering Algorithm on EtoBox?

Adaptive Parallel Tempering Algorithm by Miasojedow, Blazej; Moulines, Eric; Vihola, Matti is a scholarly article available to read on EtoBox.

What is Adaptive Parallel Tempering Algorithm about?

Parallel tempering is a generic Markov chain Monte Carlo sampling method which allows good mixing with multimodal target distributions, where conventional Metropolis-Hastings algorithms often fail. The mixing properties of the sampler depend strongly on the choice of tuning parameters, such as the temperature schedule and the proposal distribution used for local exploration. We propose an adaptive algorithm which tunes both the temperature schedule and the parameters of the random-walk Metropolis kernel automatically. We prove the convergence of the adaptation and a strong law of large numbers for the algorithm. We illustrate the performance of our method with examples. Our empirical findings indicate that the algorithm can cope well with different kind of scenarios without prior tuning.

Author

Miasojedow, Blazej; Moulines, Eric; Vihola, Matti

Published

2012

Language

EN