Entropic Independence II: Optimal Sampling and Concentration via Restricted Modified Log-Sobolev Inequalities

11/05/2021
by   Nima Anari, et al.
0

We introduce a framework for obtaining tight mixing times for Markov chains based on what we call restricted modified log-Sobolev inequalities. Modified log-Sobolev inequalities (MLSI) quantify the rate of relative entropy contraction for the Markov operator, and are notoriously difficult to establish. However, infinitesimally close to stationarity, entropy contraction becomes equivalent to variance contraction, a.k.a. a Poincare inequality, which is significantly easier to establish through, e.g., spectral analysis. Motivated by this observation, we study restricted modified log-Sobolev inequalities that guarantee entropy contraction not for all starting distributions, but for those in a large neighborhood of the stationary distribution. We show how to sample from the hardcore and Ising models on n-node graphs that have a constant δ relative gap to the tree-uniqueness threshold, in nearly-linear time O_δ(n). Notably, our bound does not depend on the maximum degree Δ, and is therefore optimal even for high-degree graphs. This improves on prior mixing time bounds of O_δ, Δ(n) and O_δ(n^2), established via (non-restricted) modified log-Sobolev and Poincare inequalities respectively. We further show that optimal concentration inequalities can still be achieved from the restricted form of modified log-Sobolev inequalities. To establish restricted entropy contraction, we extend the entropic independence framework of Anari, Jain, Koehler, Pham, and Vuong to the setting of distributions that are spectrally independent under a restricted set of external fields. We also develop an orthogonal trick that might be of independent interest: utilizing Bernoulli factories we show how to implement Glauber dynamics updates on high-degree graphs in O(1) time, assuming standard adjacency array representation of the graph.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/12/2021

On Mixing of Markov Chains: Coupling, Spectral Independence, and Entropy Factorization

For general spin systems, we prove that a contractive coupling for any l...
research
07/02/2023

Rapid mixing of global Markov chains via spectral independence: the unbounded degree case

We consider spin systems on general n-vertex graphs of unbounded degree ...
research
07/19/2023

Universality of Spectral Independence with Applications to Fast Mixing in Spin Glasses

We study Glauber dynamics for sampling from discrete distributions μ on ...
research
11/07/2021

Fast sampling via spectral independence beyond bounded-degree graphs

Spectral independence is a recently-developed framework for obtaining sh...
research
08/08/2023

Enhancing Markov and Chebyshev's inequalities

The idea of the restricted mean has been used to establish a significant...
research
09/13/2021

Entropy inequalities for random walks and permutations

We consider a new functional inequality controlling the rate of relative...

Please sign up or login with your details

Forgot password? Click here to reset