Counterdiabatic Hamiltonian Monte Carlo
Reuben Cohn-Gordon, Uroš Seljak, Dries Sels · 2026
Read Counterdiabatic Hamiltonian Monte Carlo by Reuben Cohn-Gordon, Uroš Seljak, Dries Sels (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-2602-21272.txt.
Description
Hamiltonian Monte Carlo samples from complex distributions but crawls on multimodal problems. Reuben Cohn-Gordon, Uros Seljak, and Dries Sels accelerate it in this 2026 arXiv paper with Counterdiabatic Hamiltonian Monte Carlo. Running with a time-varying Hamiltonian that interpolates from a tractable start to the target, plus a weighting scheme to remove bias, frames the method as sequential Monte Carlo sampling.
The catch is efficiency: slow interpolation is normally required. A learned counterdiabatic term added to the Hamiltonian, inspired by quantum state preparation, allows faster interpolation through a more efficient kernel. The paper relates the method to recent proposals using learned drift terms and demonstrates it on simple benchmarks. Samplers get a principled speed-up for hard multimodal targets.
Catalog record
| CREATOR | Reuben Cohn-Gordon, Uroš Seljak, Dries Sels |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 31 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-21272