Computationally Efficient Estimation of the Spectral Gap of a Markov Chain - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Proceedings of the ACM on Measurement and Analysis of Computing Systems Année : 2019

Computationally Efficient Estimation of the Spectral Gap of a Markov Chain

Résumé

We consider the problem of estimating from sample paths the absolute spectral gap 1 - λ of a reversible, irreducible and aperiodic Markov chain (X_t)_t∈N over a finite state space Ω. We propose the UCPI (Upper Confidence Power Iteration) algorithm for this problem, a low-complexity algorithm which estimates the spectral gap in time O(n) and memory space O ((łn n)^2) given n samples. This is in stark contrast with most known methods which require at least memory space O(|Ω|), so that they cannot be applied to large state spaces. We also analyze how n should scale to reach a target estimation error. Furthermore, UCPI is amenable to parallel implementation.

Dates et versions

hal-03942606 , version 1 (17-01-2023)

Identifiants

Citer

Richard Combes, Mikael Touati. Computationally Efficient Estimation of the Spectral Gap of a Markov Chain. Proceedings of the ACM on Measurement and Analysis of Computing Systems , 2019, 3 (1), pp.1-21. ⟨10.1145/3322205.3311078⟩. ⟨hal-03942606⟩
16 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More