AI RESEARCH
True Self-Avoiding Walk for Accelerating Markov-Chain Monte Carlo Integration
arXiv CS.LG
•
ArXi:2605.30532v1 Announce Type: cross We study true self-avoiding walk (TSAW) as a mechanism for improving empirical integral estimation via Marko chain Monte Carlo (MCMC). We consider finite-state adaptive sampling dynamics associated with an irreducible Marko kernel $P$ on a finite set, with stationary distribution $\pi$, in which the transition probabilities are penalized according to empirical overuse.