Distributed and Adaptive Darting Monte Carlo through Regenerations

Sungjin Ahn, Yutian Chen, Max Welling
Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics, PMLR 31:108-116, 2013.

Abstract

Darting Monte Carlo (DMC) is a MCMC procedure designed to effectively mix between multiple modes of a probability distribution. We propose an adaptive and distributed version of this method by using regenerations. This allows us to run multiple chains in parallel and adapt the shape of the jump regions as well as all other aspects of the Markov chain on the fly. We show that this significantly improves the performance of DMC because 1) a population of chains has a higher chance of finding the modes in the distribution, 2) jumping between modes becomes easier due to the adaptation of their shapes, 3) computation is much more efficient due to parallelization across multiple processors. While the curse of dimensionality is a challenge for both DMC and regeneration, we find that their combination ameliorates this issue slightly.

Cite this Paper


BibTeX
@InProceedings{pmlr-v31-ahn13a, title = {Distributed and Adaptive Darting Monte Carlo through Regenerations}, author = {Ahn, Sungjin and Chen, Yutian and Welling, Max}, booktitle = {Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics}, pages = {108--116}, year = {2013}, editor = {Carvalho, Carlos M. and Ravikumar, Pradeep}, volume = {31}, series = {Proceedings of Machine Learning Research}, address = {Scottsdale, Arizona, USA}, month = {29 Apr--01 May}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v31/ahn13a.pdf}, url = {https://proceedings.mlr.press/v31/ahn13a.html}, abstract = {Darting Monte Carlo (DMC) is a MCMC procedure designed to effectively mix between multiple modes of a probability distribution. We propose an adaptive and distributed version of this method by using regenerations. This allows us to run multiple chains in parallel and adapt the shape of the jump regions as well as all other aspects of the Markov chain on the fly. We show that this significantly improves the performance of DMC because 1) a population of chains has a higher chance of finding the modes in the distribution, 2) jumping between modes becomes easier due to the adaptation of their shapes, 3) computation is much more efficient due to parallelization across multiple processors. While the curse of dimensionality is a challenge for both DMC and regeneration, we find that their combination ameliorates this issue slightly. } }
Endnote
%0 Conference Paper %T Distributed and Adaptive Darting Monte Carlo through Regenerations %A Sungjin Ahn %A Yutian Chen %A Max Welling %B Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2013 %E Carlos M. Carvalho %E Pradeep Ravikumar %F pmlr-v31-ahn13a %I PMLR %P 108--116 %U https://proceedings.mlr.press/v31/ahn13a.html %V 31 %X Darting Monte Carlo (DMC) is a MCMC procedure designed to effectively mix between multiple modes of a probability distribution. We propose an adaptive and distributed version of this method by using regenerations. This allows us to run multiple chains in parallel and adapt the shape of the jump regions as well as all other aspects of the Markov chain on the fly. We show that this significantly improves the performance of DMC because 1) a population of chains has a higher chance of finding the modes in the distribution, 2) jumping between modes becomes easier due to the adaptation of their shapes, 3) computation is much more efficient due to parallelization across multiple processors. While the curse of dimensionality is a challenge for both DMC and regeneration, we find that their combination ameliorates this issue slightly.
RIS
TY - CPAPER TI - Distributed and Adaptive Darting Monte Carlo through Regenerations AU - Sungjin Ahn AU - Yutian Chen AU - Max Welling BT - Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics DA - 2013/04/29 ED - Carlos M. Carvalho ED - Pradeep Ravikumar ID - pmlr-v31-ahn13a PB - PMLR DP - Proceedings of Machine Learning Research VL - 31 SP - 108 EP - 116 L1 - http://proceedings.mlr.press/v31/ahn13a.pdf UR - https://proceedings.mlr.press/v31/ahn13a.html AB - Darting Monte Carlo (DMC) is a MCMC procedure designed to effectively mix between multiple modes of a probability distribution. We propose an adaptive and distributed version of this method by using regenerations. This allows us to run multiple chains in parallel and adapt the shape of the jump regions as well as all other aspects of the Markov chain on the fly. We show that this significantly improves the performance of DMC because 1) a population of chains has a higher chance of finding the modes in the distribution, 2) jumping between modes becomes easier due to the adaptation of their shapes, 3) computation is much more efficient due to parallelization across multiple processors. While the curse of dimensionality is a challenge for both DMC and regeneration, we find that their combination ameliorates this issue slightly. ER -
APA
Ahn, S., Chen, Y. & Welling, M.. (2013). Distributed and Adaptive Darting Monte Carlo through Regenerations. Proceedings of the Sixteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 31:108-116 Available from https://proceedings.mlr.press/v31/ahn13a.html.

Related Material