Entropic Mirror Monte Carlo
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914303886491648 |
|---|---|
| author | Cherradi, Anas Janati, Yazid Durmus, Alain Corff, Sylvain Le Petetin, Yohan Stoehr, Julien |
| author_facet | Cherradi, Anas Janati, Yazid Durmus, Alain Corff, Sylvain Le Petetin, Yohan Stoehr, Julien |
| contents | Importance sampling is a Monte Carlo method which designs estimators of expectations under a target distribution using weighted samples from a proposal distribution. When the target distribution is complex, such as multimodal distributions in highdimensional spaces, the efficiency of importance sampling critically depends on the choice of the proposal distribution. In this paper, we propose a novel adaptive scheme for the construction of efficient proposal distributions. Our algorithm promotes efficient exploration of the target distribution by combining global sampling mechanisms with a delayed weighting procedure. The proposed weighting mechanism plays a key role by enabling rapid resampling in regions where the proposal distribution is poorly adapted to the target. Our sampling algorithm is shown to be geometrically convergent under mild assumptions and is illustrated through various numerical experiments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_03165 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Entropic Mirror Monte Carlo Cherradi, Anas Janati, Yazid Durmus, Alain Corff, Sylvain Le Petetin, Yohan Stoehr, Julien Methodology Machine Learning Importance sampling is a Monte Carlo method which designs estimators of expectations under a target distribution using weighted samples from a proposal distribution. When the target distribution is complex, such as multimodal distributions in highdimensional spaces, the efficiency of importance sampling critically depends on the choice of the proposal distribution. In this paper, we propose a novel adaptive scheme for the construction of efficient proposal distributions. Our algorithm promotes efficient exploration of the target distribution by combining global sampling mechanisms with a delayed weighting procedure. The proposed weighting mechanism plays a key role by enabling rapid resampling in regions where the proposal distribution is poorly adapted to the target. Our sampling algorithm is shown to be geometrically convergent under mild assumptions and is illustrated through various numerical experiments. |
| title | Entropic Mirror Monte Carlo |
| topic | Methodology Machine Learning |
| url | https://arxiv.org/abs/2602.03165 |