Metraipolis › archive › dc_fc985c51f733
Foresight: Learning Twisted SMC for Rare Events from Reversed Relaxations — arXiv:2611.00903
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.00903
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T11:20:48.572Z
- SHA-256 of the archived text
93578997bcbfef6143db6530ffd91807cd167b3414de4036699a202e38ffec57- Stable URL of this page
https://metraipolis.com/documento/dc_fc985c51f733
How to cite
Foresight: Learning Twisted SMC for Rare Events from Reversed Relaxations — arXiv:2611.00903 — arXiv (en), https://arxiv.org/abs/2611.00903 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fc985c51f733 (Metraipolis archive dc_fc985c51f733, sha256 93578997bcbfef6143db6530ffd91807cd167b3414de4036699a202e38ffec57)
BibTeX
@misc{dcfc985c51f733,
title = {Foresight: Learning Twisted SMC for Rare Events from Reversed Relaxations — arXiv:2611.00903},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.00903},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fc985c51f733; sha256 93578997bcbfef6143db6530ffd91807cd167b3414de4036699a202e38ffec57},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.00903. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We consider the probability that a reversible Markov chain started in a set $A$ is in a rare set $B$ after a fixed horizon, a quantity often needed for many start laws, down to single states. Twisted sequential Monte Carlo (SMC) estimates such probabilities without bias, with small variance when its proposal and twist approximate the dynamics conditioned on reaching $B$, but these components are usually learned from forward simulations, which contain almost no successful paths when the event is rare. We introduce Foresight-SMC, which learns both components from the target side: under reversibility, relaxations started in $B$ that return to $A$ are, read backwards, exact samples of the conditioned paths, and their number is governed by the return probability rather than by the rare probability. Extending this reversal to the recorded propose--accept decisions of a Metropolis-adjusted sampler, we fit a controlled Langevin proposal with exact likelihood ratios by maximum likelihood, together with a classifier twist; the resulting estimator is unbiased for any start law. A variance analysis shows that resampling reduces the variance when the importance weights degenerate, as they do over long horizons. On an Allen--Cahn field, the learned proposal alone suffices, and one trained model is more accurate than splitting at equal cost. On alanine tripeptide, resampling with the learned twist is necessary, and Foresight-SMC needs less computation per query than adaptive multilevel splitting on localized start sets. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2611.00903 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-03T11:02:37.480Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).