Preconditioning Hamiltonian Monte Carlo by minimizing Fisher Divergence
Adrian Seyboldt, Eliot L. Carlson, Bob Carpenter · 2026
Read Preconditioning Hamiltonian Monte Carlo by minimizing Fisher Divergence by Adrian Seyboldt, Eliot L. Carlson, Bob Carpenter (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-2603-18845.txt.
Description
Seyboldt, Carlson, and Carpenter improve Hamiltonian Monte Carlo by picking better preconditioners. HMC scales well with dimension, but target curvature sets a large constant cost. Standard practice estimates variance from warmup draws, which converges slowly. Their estimator instead shrinks the Fisher divergence between a linearly reshaped density and a standard normal. Three forms are offered: diagonal, dense, and low-rank plus diagonal. On 114 models from posteriordb, the diagonal form beats Stan and PyMC defaults by a median factor of 1.3. The low-rank form wins by a median factor of 4.
Catalog record
| CREATOR | Adrian Seyboldt, Eliot L. Carlson, Bob Carpenter |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 75049 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.18845