Ensemble Slice Sampling. (arXiv:2002.06212v1 [stat.ML])
<a href="http://arxiv.org/find/stat/1/au:+Karamanis_M/0/1/0/all/0/1">Minas Karamanis</a>, <a href="http://arxiv.org/find/stat/1/au:+Beutler_F/0/1/0/all/0/1">Florian Beutler</a>
Slice Sampling has emerged as a powerful Markov Chain Monte Carlo algorithm
that adapts to the characteristics of the target distribution with minimal
hand-tuning. However, Slice Sampling’s performance is highly sensitive to the
user-specified initial length scale hyperparameter. Moreover, Slice Sampling
generally struggles with poorly scaled or strongly correlated distributions.
This paper introduces Ensemble Slice Sampling, a new class of algorithms that
bypasses such difficulties by adaptively tuning the length scale. Furthermore,
Ensemble Slice Sampling’s performance is immune to linear correlations by
exploiting an ensemble of parallel walkers. These algorithms are trivial to
construct, require no hand-tuning, and can easily be implemented in parallel
computing environments. Empirical tests show that Ensemble Slice Sampling can
improve efficiency by more than an order of magnitude compared to conventional
MCMC methods on highly correlated target distributions such as the
Autoregressive Process of Order 1 and the Correlated Funnel distribution.
Slice Sampling has emerged as a powerful Markov Chain Monte Carlo algorithm
that adapts to the characteristics of the target distribution with minimal
hand-tuning. However, Slice Sampling’s performance is highly sensitive to the
user-specified initial length scale hyperparameter. Moreover, Slice Sampling
generally struggles with poorly scaled or strongly correlated distributions.
This paper introduces Ensemble Slice Sampling, a new class of algorithms that
bypasses such difficulties by adaptively tuning the length scale. Furthermore,
Ensemble Slice Sampling’s performance is immune to linear correlations by
exploiting an ensemble of parallel walkers. These algorithms are trivial to
construct, require no hand-tuning, and can easily be implemented in parallel
computing environments. Empirical tests show that Ensemble Slice Sampling can
improve efficiency by more than an order of magnitude compared to conventional
MCMC methods on highly correlated target distributions such as the
Autoregressive Process of Order 1 and the Correlated Funnel distribution.
http://arxiv.org/icons/sfx.gif
