Condensed Matter > Disordered Systems and Neural Networks
[Submitted on 24 Apr 2012 (this version), latest version 10 Aug 2012 (v2)]
Title:Mean-field theory for the inverse Ising problem at low temperatures
View PDFAbstract:Taking a set of spin configurations sampled from the equilibrium distribution of an Ising model, can the underlying couplings between spins be reconstructed from a large number of such samples? This inverse Ising problem is a paradigmatic inverse problem with applications in neural biology, protein structure determination and gene expression analysis. Typically a large number of spins (representing neurons, genetic loci) is involved, as well as a large number of interactions between them. Mean-field approximations are often used to invert the relationship between the model parameters (external fields and couplings between spins) and observables (magnetisations and correlations), allowing to determine model parameters from data. However, all known mean-field methods fail at low temperatures. In this note, we show how clustering spin configurations can approximate thermodynamic states, and how mean-field methods applied to these thermodynamic states allow to reconstruct Ising models also at low temperatures.
Submission history
From: H. Chau Nguyen [view email][v1] Tue, 24 Apr 2012 13:39:55 UTC (33 KB)
[v2] Fri, 10 Aug 2012 14:28:23 UTC (35 KB)
Current browse context:
cond-mat.dis-nn
Change to browse by:
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
IArxiv Recommender
(What is IArxiv?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.