Strong Self-Concordance and Sampling

11/13/2019
by   Aditi Laddha, et al.
0

Motivated by the Dikin walk, we develop aspects of an interior-point theory for sampling in high dimension. Specifically, we introduce symmetric and strong self-concordance. These properties imply that the corresponding Dikin walk mixes in Õ(nν̅) steps from a warm start in a convex body in R^n using a strongly self-concordant barrier with symmetric self-concordance parameter ν̅. For many natural barriers, ν̅ is roughly bounded by ν, the standard self-concordance parameter. We show that this property and strong self-concordance hold for the Lee-Sidford barrier. As a consequence, we obtain the first walk to mix in Õ(n^2) steps for an arbitrary polytope in R^n. Strong self-concordance for other barriers leads to an interesting (and unexpected) connection — for the universal and entropic barriers, it is implied by the KLS conjecture.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/21/2021

The entropic barrier is n-self-concordant

For any convex body K ⊆ℝ^n, S. Bubeck and R. Eldan introduced the entrop...
research
10/23/2017

Fast MCMC sampling algorithms on polytopes

We propose and analyze two new MCMC sampling algorithms, the Vaidya walk...
research
03/06/2018

John's Walk

We present an affine-invariant random walk for drawing uniform random sa...
research
02/06/2023

RoboCup 2022 AdultSize Winner NimbRo: Upgraded Perception, Capture Steps Gait and Phase-based In-walk Kicks

Beating the human world champions by 2050 is an ambitious goal of the Hu...
research
03/01/2023

Sampling with Barriers: Faster Mixing via Lewis Weights

We analyze Riemannian Hamiltonian Monte Carlo (RHMC) for sampling a poly...
research
07/24/2023

Efficiently Sampling the PSD Cone with the Metric Dikin Walk

Semi-definite programs represent a frontier of efficient computation. Wh...
research
01/06/2022

No self-concordant barrier interior point method is strongly polynomial

It is an open question to determine if the theory of self-concordant bar...

Please sign up or login with your details

Forgot password? Click here to reset