Concentration inequalities for polynomials of contracting Ising models
read the original abstract
We study the concentration of a degree-$d$ polynomial of the $N$ spins of a general Ising model, in the regime where single-site Glauber dynamics is contracting. For $d=1$, Gaussian concentration was shown by Marton (1996) and Samson (2000) as a special case of concentration for convex Lipschitz functions, and extended to a variety of related settings by e.g., Chazottes et al. (2007) and Kontorovich and Ramanan (2008). For $d=2$, exponential concentration was shown by Marton (2003) on lattices. We treat a general fixed degree $d$ with $O(1)$ coefficients, and show that the polynomial has variance $O(N^d)$ and, after rescaling it by $N^{-d/2}$, its tail probabilities decay as $\exp(- c\, r^{2/d})$ for deviations of $r \geq C \log N$.
This paper has not been read by Pith yet.
Forward citations
Cited by 1 Pith paper
-
Learning from weakly dependent data under Dobrushin's condition
Generalization and learnability bounds for hypothesis classes under Dobrushin's condition on weakly dependent data, with degradation by only constant or log factors relative to i.i.d. settings.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.