Presentation 2006-10-11
On Relation between Exchange Ratio and Kullback Divergence in Exchange Monte Carlo Method
Kenji NAGATA, Sumio WATANABE,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The exchange Monte Carlo method was proposed as an improved algorithm of Markov Chain Monte Carlo method, and its effectiveness has been shown in many fields. In the exchange Monte Carlo method, the setting of temperatures is important to make the algorithm efficient because this setting controls the exchange ratio, with which the position exchange between two sequences is accepted. The symmetrized Kullback divergence between two distributions with different temperatures is used as the criterion for setting of temperature. However, the mathmatical relation between the symmetrized Kullback divergence and the exchange ratio is not clarified. In this paper, we analytically calculate the asymptotic form of the symmetirized Kullback divergence and the exchange ratio for the arbitrary distribution, and clarify the relation between the symmetrized Kullback divergence and the exchange ratio.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Exchange Monte Carlo Method / Kullback Divergence / Exchange Ratio
Paper # NC2006-52
Date of Issue

Conference Information
Committee NC
Conference Date 2006/10/4(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) On Relation between Exchange Ratio and Kullback Divergence in Exchange Monte Carlo Method
Sub Title (in English)
Keyword(1) Exchange Monte Carlo Method
Keyword(2) Kullback Divergence
Keyword(3) Exchange Ratio
1st Author's Name Kenji NAGATA
1st Author's Affiliation Department of Computer Science Tokyo Institute of Technology()
2nd Author's Name Sumio WATANABE
2nd Author's Affiliation PI Lab., Tokyo Institute of Technology
Date 2006-10-11
Paper # NC2006-52
Volume (vol) vol.106
Number (no) 279
Page pp.pp.-
#Pages 6
Date of Issue