Data quality for the inverse Ising problem

Aurélien Decelle 1 Federico Ricci-Tersenghi 2 Pan Zhang 3
1 TAO - Machine Learning and Optimisation
LRI - Laboratoire de Recherche en Informatique, UP11 - Université Paris-Sud - Paris 11, Inria Saclay - Ile de France, CNRS - Centre National de la Recherche Scientifique : UMR8623
Abstract : There are many methods proposed for inferring parameters of the Ising model from given data, that is a set of configurations generated according to the model itself. However little attention has been paid until now to the data, e.g. how the data is generated, whether the inference error using one set of data could be smaller than using another set of data, etc. In this paper we address the data quality problem in the kinetic inverse Ising problem. We quantify the quality of data using effective rank of the correlation matrix, and show that data gathered in a out of-equilibrium regime has a better quality than data gathered in equilibrium for coupling reconstruction. We also propose a matrix-perturbation based method for tuning the quality of given data and for removing bad-quality (i.e. redundant) configurations from data.
Liste complète des métadonnées

Littérature citée [21 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01250822
Contributeur : Aurélien Decelle <>
Soumis le : mardi 5 janvier 2016 - 12:58:39
Dernière modification le : jeudi 7 février 2019 - 16:50:04
Document(s) archivé(s) le : jeudi 7 avril 2016 - 15:16:49

Fichier

1511.01190v1.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Citation

Aurélien Decelle, Federico Ricci-Tersenghi, Pan Zhang. Data quality for the inverse Ising problem. Journal of Physics A: Mathematical and Theoretical, IOP Publishing, 2016, 49 (38), 〈10.1088/1751-8113/49/38/384001〉. 〈hal-01250822〉

Partager

Métriques

Consultations de la notice

653

Téléchargements de fichiers

189