Adaptive Pseudo-Marginal Algorithm
Sarra Abaoubida, Mylène Bédard, Florian Maire · 2025
Read Adaptive Pseudo-Marginal Algorithm by Sarra Abaoubida, Mylène Bédard, Florian Maire (2025) 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-2509-24820.txt.
Description
Sarra Abaoubida, Mylène Bédard, and Florian Maire automate a tuning chore in Bayesian sampling in a 2025 statistics paper. The pseudo-marginal algorithm samples from distributions whose density can only be estimated, not computed. Its accuracy rides on one knob: how many particles feed each estimate. More particles mean better estimates and slower runs, and picking the sweet spot by hand takes trial and error.
Their adaptive version turns the knob itself as the chain runs, steering toward the best value. Convergence holds under stated conditions. On two tests, one a real study of lung infection in preschool children, the self-tuning sampler matched or beat the hand-tuned original. Removing babysitting from a finicky sampler widens its use.
Catalog record
| CREATOR | Sarra Abaoubida, Mylène Bédard, Florian Maire |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 110 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-24820