Concentration of Multilinear Functions of the Ising Model with Applications to Network Data

Abstract

We prove near-tight concentration of measure for polynomial functions of the Ising model under high temperature. For any degree $d$, we show that a degree-$d$ polynomial of a $n$-spin Ising model exhibits exponential tails that scale as $\exp (-r^{2/d})$ at radius $r={\tilde{\Omega }}_d(n^{d/2})$. Our concentration radius is optimal up to logarithmic factors for constant $d$, improving known results by polynomial factors in the number of spins. We demonstrate the efficacy of polynomial functions as statistics for testing the strength of interactions in social networks in both synthetic and real world data.

arXiv commentTo appear in NIPS 2017

Cited by in corpus (5)

• Higher order concentration for functions of weakly dependent random variables
• Optimal Single Sample Tests for Structured versus Unstructured Network Data
• Modified log-Sobolev inequalities, Beckner inequalities and moment estimates
• Privately Learning Markov Random Fields
• Regression from Dependent Observations