Optimizing Markov chain partitions with Bregman divergences, animated by Claude
michaelchchoi · x · 2026-10-23
Researcher michaelchchoi demonstrates using Bregman divergences and information projections to optimize partition selection for accelerating Markov chains. On a 50x50 grid with k=10 blocks, he compares three schemes: lazy Metropolis as baseline, 9 lightest states as singleton blocks with the rest in one block, and balanced blocks. The comparison animation was created by Claude.
More from Research
- A classic world model (s, a → s′) acting as a neural MuJoCo simulator — yshan2u · 2026-10-23
- Gen-HumanWholeBody: open 600-hour whole-body manipulation dataset for robots — AdinaYakup · 2026-10-23
- DynaVelo, a neural ODE model of single-cell dynamics, published in Nature Biotechnology — anshulkundaje · 2026-10-23
- 120-case memory benchmark: Hy-Memory hits 100% recall but fails all 44 empty-output cases — stevyhacker · 2026-10-23
- SplatLab: an autonomous AI scientist that runs 3DGS research end to end — UCSanDiego · 2026-10-23
- RTAformer tackles 'relational semantic inertia' in video scene graph generation — Jaeyeon Heo · 2026-10-23