arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.09170stat.MLcs.AIcs.LG
收藏

量化扩散模型的先验不确定性

Quantifying Epistemic Uncertainty in Diffusion Models

Berkeley Lab & ICSI(伯克利实验室与ICS实验室);UC Berkeley & ICSI(加州大学伯克利分校与ICS实验室);Berkeley Lab(伯克利实验室);New York University(纽约大学)

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

AI总结:

本文提出FLARE方法,通过基于信息量的随机近似技术,有效分离扩散模型的先验不确定性,提升生成数据的可信度评估。

摘要:

为确保高质量输出,量化扩散模型的先验不确定性至关重要。现有方法往往不可靠,因为它们混合了先验和偶然不确定性。我们介绍了一种基于信息量的 方法,该方法明确分离了先验方差,从而为生成数据产生更可靠的可信度分数。为使该方法可扩展,我们提出了FLARE(Fisher-Laplace Randomized Estimator),它使用均匀随机的模型参数子集来近似信息量。实证上,FLARE在合成时间序列生成任务中提高了不确定性估计,比其他方法实现了更准确和可靠的过滤。理论上,我们界定了随机近似的收敛率,并提供了分析和实证证据表明,最后一层Laplace近似不足以完成此任务。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑