Anytime marginal MAP inference

Denis Deratani Mauá; C.P. de Campos

Anytime marginal MAP inference

Denis Deratani Mauá, C.P. de Campos

Research output: Chapter in Book/Report/Conference proceeding › Chapter › Academic › peer-review

Abstract

This paper presents a new anytime algorithm for the marginal MAP problem in graphical models of bounded treewidth. We show asymptotic convergence and theoretical error bounds for any fixed step. Experiments show that it compares well to a state-of-the-art systematic search algorithm.

Original language	English
Title of host publication	International Conference on Machine Learning (ICML)
Publisher	Omnipress
Pages	1471-1478
Number of pages	8
Publication status	Published - 2012
Externally published	Yes
Event	29th International Conference on Machine Learning (ICML 2012) - Edinburgh, United Kingdom Duration: 26 Jun 2012 → 1 Jul 2012 Conference number: 29

Conference

Conference	29th International Conference on Machine Learning (ICML 2012)
Abbreviated title	ICML 2012
Country/Territory	United Kingdom
City	Edinburgh
Period	26/06/12 → 1/07/12

Bibliographical note

(oral presentation, double-blind peer reviewed by >3 reviewers)

Cite this

@inbook{981eb3c2e86649f0ae81aae93f1226b6,

title = "Anytime marginal MAP inference",

abstract = "This paper presents a new anytime algorithm for the marginal MAP problem in graphical models of bounded treewidth. We show asymptotic convergence and theoretical error bounds for any fixed step. Experiments show that it compares well to a state-of-the-art systematic search algorithm.",

author = "Mau{\'a}, {Denis Deratani} and {de Campos}, C.P.",

note = "(oral presentation, double-blind peer reviewed by >3 reviewers); 29th International Conference on Machine Learning (ICML 2012), ICML 2012 ; Conference date: 26-06-2012 Through 01-07-2012",

year = "2012",

language = "English",

pages = "1471--1478",

booktitle = "International Conference on Machine Learning (ICML)",

publisher = "Omnipress",

}

TY - CHAP

T1 - Anytime marginal MAP inference

AU - Mauá, Denis Deratani

AU - de Campos, C.P.

N1 - Conference code: 29

PY - 2012

Y1 - 2012

N2 - This paper presents a new anytime algorithm for the marginal MAP problem in graphical models of bounded treewidth. We show asymptotic convergence and theoretical error bounds for any fixed step. Experiments show that it compares well to a state-of-the-art systematic search algorithm.

AB - This paper presents a new anytime algorithm for the marginal MAP problem in graphical models of bounded treewidth. We show asymptotic convergence and theoretical error bounds for any fixed step. Experiments show that it compares well to a state-of-the-art systematic search algorithm.

M3 - Chapter

SP - 1471

EP - 1478

BT - International Conference on Machine Learning (ICML)

PB - Omnipress

T2 - 29th International Conference on Machine Learning (ICML 2012)

Y2 - 26 June 2012 through 1 July 2012

ER -

Anytime marginal MAP inference

Abstract

Conference

Bibliographical note

Fingerprint

Cite this