Generalized Darting Monte Carlo

Cristian Sminchisescu, Max Welling
; Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics, PMLR 2:516-523, 2007.

Abstract

One of the main shortcomings of Markov chain Monte Carlo samplers is their inability to mix between modes of the target distribution. In this paper we show that advance knowledge of the location of these modes can be incorporated into the MCMC sampler by introducing mode-hopping moves that satisfy detailed balance. The proposed sampling algorithm explores local mode structure through local MCMC moves (e.g. diffusion or Hybrid Monte Carlo) but in addition also represents the relative strengths of the different modes correctly using a set of global moves. This ‘mode-hopping’ MCMC sampler can be viewed as a generalization of the darting method [1]. We illustrate the method on a ‘real world’ vision application of inferring 3-D human body pose from single 2-D images.

Cite this Paper


BibTeX
@InProceedings{pmlr-v2-sminchisescu07a, title = {Generalized Darting Monte Carlo}, author = {Cristian Sminchisescu and Max Welling}, booktitle = {Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics}, pages = {516--523}, year = {2007}, editor = {Marina Meila and Xiaotong Shen}, volume = {2}, series = {Proceedings of Machine Learning Research}, address = {San Juan, Puerto Rico}, month = {21--24 Mar}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v2/sminchisescu07a/sminchisescu07a.pdf}, url = {http://proceedings.mlr.press/v2/sminchisescu07a.html}, abstract = {One of the main shortcomings of Markov chain Monte Carlo samplers is their inability to mix between modes of the target distribution. In this paper we show that advance knowledge of the location of these modes can be incorporated into the MCMC sampler by introducing mode-hopping moves that satisfy detailed balance. The proposed sampling algorithm explores local mode structure through local MCMC moves (e.g. diffusion or Hybrid Monte Carlo) but in addition also represents the relative strengths of the different modes correctly using a set of global moves. This ‘mode-hopping’ MCMC sampler can be viewed as a generalization of the darting method [1]. We illustrate the method on a ‘real world’ vision application of inferring 3-D human body pose from single 2-D images.} }
Endnote
%0 Conference Paper %T Generalized Darting Monte Carlo %A Cristian Sminchisescu %A Max Welling %B Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2007 %E Marina Meila %E Xiaotong Shen %F pmlr-v2-sminchisescu07a %I PMLR %J Proceedings of Machine Learning Research %P 516--523 %U http://proceedings.mlr.press %V 2 %W PMLR %X One of the main shortcomings of Markov chain Monte Carlo samplers is their inability to mix between modes of the target distribution. In this paper we show that advance knowledge of the location of these modes can be incorporated into the MCMC sampler by introducing mode-hopping moves that satisfy detailed balance. The proposed sampling algorithm explores local mode structure through local MCMC moves (e.g. diffusion or Hybrid Monte Carlo) but in addition also represents the relative strengths of the different modes correctly using a set of global moves. This ‘mode-hopping’ MCMC sampler can be viewed as a generalization of the darting method [1]. We illustrate the method on a ‘real world’ vision application of inferring 3-D human body pose from single 2-D images.
RIS
TY - CPAPER TI - Generalized Darting Monte Carlo AU - Cristian Sminchisescu AU - Max Welling BT - Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics PY - 2007/03/11 DA - 2007/03/11 ED - Marina Meila ED - Xiaotong Shen ID - pmlr-v2-sminchisescu07a PB - PMLR SP - 516 DP - PMLR EP - 523 L1 - http://proceedings.mlr.press/v2/sminchisescu07a/sminchisescu07a.pdf UR - http://proceedings.mlr.press/v2/sminchisescu07a.html AB - One of the main shortcomings of Markov chain Monte Carlo samplers is their inability to mix between modes of the target distribution. In this paper we show that advance knowledge of the location of these modes can be incorporated into the MCMC sampler by introducing mode-hopping moves that satisfy detailed balance. The proposed sampling algorithm explores local mode structure through local MCMC moves (e.g. diffusion or Hybrid Monte Carlo) but in addition also represents the relative strengths of the different modes correctly using a set of global moves. This ‘mode-hopping’ MCMC sampler can be viewed as a generalization of the darting method [1]. We illustrate the method on a ‘real world’ vision application of inferring 3-D human body pose from single 2-D images. ER -
APA
Sminchisescu, C. & Welling, M.. (2007). Generalized Darting Monte Carlo. Proceedings of the Eleventh International Conference on Artificial Intelligence and Statistics, in PMLR 2:516-523

Related Material