TU Berlin

Methods of Artificial IntelligencePublications

Page Content

to Navigation

List of Publications

A statistical physics approach to learning curves for the inverse Ising problem
Citation key 1742-5468-2017-6-063406
Author Ludovica Bachschmid-Romano and Manfred Opper
Pages 063406
Year 2017
ISSN 1742-5468
DOI https://doi.org/10.1088/1742-5468/aa727d
Journal Journal of Statistical Mechanics: Theory and Experiment
Volume 2017
Number 6
Abstract Using methods of statistical physics, we analyse the error of learning couplings in large Ising models from independent data (the inverse Ising problem). We concentrate on learning based on local cost functions, such as the pseudo-likelihood method for which the couplings are inferred independently for each spin. Assuming that the data are generated from a true Ising model, we compute the reconstruction error of the couplings using a combination of the replica method with the cavity approach for densely connected systems. We show that an explicit estimator based on a quadratic cost function achieves minimal reconstruction error, but requires the length of the true coupling vector as prior knowledge. A simple mean field estimator of the couplings which does not need such knowledge is asymptotically optimal, i.e. when the number of observations is much larger than the number of spins. Comparison of the theory with numerical simulations shows excellent agreement for data generated from two models with random couplings in the high temperature region: a model with independent couplings (Sherrington–Kirkpatrick model), and a model where the matrix of couplings has a Wishart distribution.
Link to publication Download Bibtex entry


Quick Access

Schnellnavigation zur Seite über Nummerneingabe