Steering Diffusion Models to Rare Events with Sequential Monte Carlo

Key points

• Diffusion models are increasingly used as surrogates for expensive simulators in weather prediction, molecular dynamics, and materials design.1
• In these models, computing the probability $p0[E]$ of an event $E$ is difficult, especially when the event of interest is rare.1
• A stable estimate using Monte Carlo becomes computationally intractable, requiring a growing sample size $\propto\!1/p0[E]$ to compensate for an increasing rarity.1
• 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.1

Sources (1)

• [1]Steering Diffusion Models to Rare Events with Sequential Monte Carlo
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 6, 04:38 PM

  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.

  Diffusion models are increasingly used as surrogates for expensive simulators in weather prediction, molecular dynamics, and materials design.

Extractive summary: sentences quoted from the sources.