New Sampling Kernel for Finite Markov Chains

michaelchchoi · x · 2026-07-17

The author announced an update to their preprint published today on arXiv, thanking two reviewers for their constructive feedback.

The paper is titled On additive averaging kernels for finite Markov chains. It investigates an additive mixing kernel:

\[ A\alpha = \alpha P + (1-\alpha)G \]

where P is the base sampler and G is the Gibbs / group-averaging kernel. The author compares this with the combination-based sampling method GPG, proposing it as an alternative approach.

Original post →

More from Research

Research channel →