Perturbed infinite-state Markov systems with holes and its application
Haruyoshi Tanaka

TL;DR
This paper studies how Gibbs measures for infinite-state Markov systems with holes behave under small perturbations, extending previous finite-state results and applying spectral analysis techniques.
Contribution
It develops a spectral gap approach for infinite-state systems and analyzes measure convergence and splitting behavior under perturbations.
Findings
Convergence of Gibbs measures as perturbation parameter tends to zero.
Representation of limiting measures in perturbed infinite-state systems.
Splitting behavior of Gibbs measures in systems with holes.
Abstract
We consider a perturbed system , where is a topological Markov shift with a countably infinite state space, and is a real-valued potential on X depending on a small parameter . We assume that for each , the system has a unique transitive component and a unique Gibbs measure (or more generally, a Ruelle-Perron-Frobenius (RPF) measure) , while the unperturbed system possesses multiple transitive components and Gibbs measures on these components. We investigate the convergence of the measure as and the representation of the limiting measure, if it exists. In previous work [T. 2020], we considered the finite state case. Our approach relies on a development of the Schur-Frobenius factorization theorem, which we apply to demonstrate a spectral gap property…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsPetri Nets in System Modeling
