Quantifying Epistemic Uncertainty in Diffusion Models

TL;DR

This paper introduces FLARE, a Fisher information-based method to accurately quantify epistemic uncertainty in diffusion models, improving reliability over existing approaches especially in synthetic time-series generation.

Contribution

The paper presents FLARE, a scalable Fisher information-based approach that isolates epistemic uncertainty, with theoretical bounds and empirical validation demonstrating its effectiveness.

Findings

FLARE outperforms existing methods in uncertainty estimation for synthetic time-series.

Theoretical bounds on the convergence rate of the randomized Fisher information approximation.

Last-layer Laplace approximations are shown to be insufficient for epistemic uncertainty quantification.

Abstract

To ensure high quality outputs, it is important to quantify the epistemic uncertainty of diffusion models. Existing methods are often unreliable because they mix epistemic and aleatoric uncertainty. We introduce a method based on Fisher information that explicitly isolates epistemic variance, producing more reliable plausibility scores for generated data. To make this approach scalable, we propose FLARE (Fisher-Laplace Randomized Estimator), which approximates the Fisher information using a uniformly random subset of model parameters. Empirically, FLARE improves uncertainty estimation in synthetic time-series generation tasks, achieving more accurate and reliable filtering than other methods. Theoretically, we bound the convergence rate of our randomized approximation and provide analytic and empirical evidence that last-layer Laplace approximations are insufficient for this task.