Optimising two-block averaging kernels to speed up Markov chains
Ryan J.Y. Lim, Michael C.H. Choi \orcid 0000-0003-0309-3217 · 2026
Read Optimising two-block averaging kernels to speed up Markov chains by Ryan J.Y. Lim, Michael C.H. Choi \orcid 0000-0003-0309-3217 (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-10318.txt.
Description
Ryan J Y Lim and Michael C H Choi study how to pick a two-block split that helps a finite Markov chain mix faster under group averaging. They link two goals, KL divergence and Frobenius distance to stationarity, to the induced projection chain. The first link gives decay rates through the log-Sobolev constant, while the second yields a Cheeger-type cut criterion. Since exact search is costly, they offer practical approximations, and numerical tests show the best cuts cut total variation distance to stationarity a lot.
Catalog record
| CREATOR | Ryan J.Y. Lim, Michael C.H. Choi \orcid 0000-0003-0309-3217 |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 101697 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.10318