Consistent inference of a general model using the pseudo-likelihood method

Abstract

Recently maximum pseudo-likelihood (MPL) inference method has been successfully applied to statistical physics models with intractable likelihoods. We use information theory to derive a relation between the pseudo-likelihood and likelihood functions. We use this relation to show consistency of the pseudo-likelihood method for a general model.

A new more mathematically rigorous version accepted for publication in Physical Review E: Rapid Communication

References in corpus (2)

• Direct-coupling analysis of residue co-evolution captures native contacts across many protein families
• Improved contact prediction in proteins: Using pseudolikelihoods to infer Potts models

Cited by in corpus (6)

• Inverse statistical problems: from the inverse Ising problem to data science
• Causal Inference Under Interference And Network Uncertainty
• A statistical physics approach to learning curves for the Inverse Ising problem
• Network reconstruction via the minimum description length principle
• Structure Learning in Inverse Ising Problems Using ${\ell }_2$-Regularized Linear Estimator
• Pseudo-likelihood produces associative memories able to generalize, even for asymmetric couplings