Pairwise MRF Calibration by Perturbation of the Bethe Reference Point

Cyril Furtlehner 1 Yufei Han 2 Jean-Marc Lasgouttes 2 Victorin Martin 2
1 TAO - Machine Learning and Optimisation
CNRS - Centre National de la Recherche Scientifique : UMR8623, Inria Saclay - Ile de France, UP11 - Université Paris-Sud - Paris 11, LRI - Laboratoire de Recherche en Informatique
Abstract : We investigate different ways of generating approximate solutions to the inverse problem of pairwise Markov random field (MRF) model learning. We focus mainly on the inverse Ising problem, but discuss also the somewhat related inverse Gaussian problem. In both cases, the belief propagation algorithm can be used in closed form to perform inference tasks. Our approach consists in taking the Bethe mean-field solution as a reference point for further perturbation procedures. We remark indeed that both the natural gradient and the best link to be added to a maximum spanning tree (MST) solution can be computed analytically. These observations open the way to many possible algorithms, able to find approximate sparse solutions compatible with belief propagation inference procedures. Experimental tests on various datasets with refined $L_0$ or $L_1$ regularization procedures indicate that this approach may be a competitive and useful alternative to existing ones.
Complete list of metadatas

Cited literature [36 references]  Display  Hide  Download

https://hal.inria.fr/hal-00743334
Contributor : Cyril Furtlehner <>
Submitted on : Thursday, October 18, 2012 - 5:02:37 PM
Last modification on : Friday, May 25, 2018 - 12:02:03 PM
Long-term archiving on : Saturday, January 19, 2013 - 3:40:51 AM

File

RR-8059.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00743334, version 1

Collections

Citation

Cyril Furtlehner, Yufei Han, Jean-Marc Lasgouttes, Victorin Martin. Pairwise MRF Calibration by Perturbation of the Bethe Reference Point. [Research Report] RR-8059, INRIA. 2012, pp.35. ⟨hal-00743334⟩

Share

Metrics

Record views

390

Files downloads

200