arXiv：cs.LG· Aavash Subedi, Tim Reichelt, Christopher Williams, Philip Stier, Yee Whye Teh, Saifuddin Syed·· 7 小时前AI 评分34
DireSMC：用序列蒙特卡洛引导扩散模型采样罕见事件
Steering Diffusion Models to Rare Events with Sequential Monte Carlo
AI 导读
研究者提出 DireSMC，一种序列蒙特卡洛方案，通过引导带权样本群体逼近罕见事件，同时给出其概率的校准估计。该方法用事件集的解析松弛构造引导，在基于分数的气候模拟器上对 10⁻³ 至 10⁻⁵ 的罕见概率取得准确估计，相比蒙特卡洛实现 9× 至 1413× 的净加速。
正文
Abstract:Diffusion models are increasingly used as surrogates for expensive simulators in weather prediction, molecular dynamics, and materials design. In these models, computing the probability $p_0[E]$ of an event $E$ is difficult, especially when the event of interest is rare. A stable estimate using Monte Carlo becomes computationally intractable, requiring a growing sample size $\propto\!1/p_0[E]$ to compensate for an increasing rarity. In this paper, we present Diffusion Importance Sampling of Rare Events or DireSMC, a sequential Monte Carlo scheme that guides a population of weighted samples towards the rare event, giving access not only to samples but also to a calibrated estimate of its probability. We set up our guidance using an analytical relaxation of the event set, allowing the method to easily extend to a wide range of user-defined rare events. We validate our method on a toy problem with analytical solutions and on a score-based climate emulator, where we obtain accurate rare-event probabilities on a range of rarities from $10^{-3}$ to $10^{-5}$, achieving net speed-ups of $9\times$ to $1413\times$ over Monte Carlo.
| Comments: | A previous version of this work was presented at the NeurIPS 2026: AI for Stochastic Dynamics workshop |
| Subjects: | Machine Learning (stat.ML); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.08652 [stat.ML] |
| (or arXiv:2610.08652v1 [stat.ML] for this version) | |
| https://doi.org/10.48550/arXiv.2610.08652 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Aavash Subedi [view email]
[v1]
Tue, 6 Oct 2026 16:38:03 UTC (9,235 KB)
来源：arXiv：cs.LG · arxiv.org