Quantifying Epistemic Uncertainty in Diffusion Models

Aditi Gupta, Raphael A Meyer, Yotam Yaniv, Elynn Chen, N. Benjamin Erichson

TL;DR: FLARE is a randomized Fisher–Laplace method that isolates epistemic uncertainty in diffusion models, producing accurate uncertainty estimates for diagnostics and filtering low-confidence samples.

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.

Submission Number: 1574