arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2411.08587cs.LGcs.AI
收藏

DeepUQ:评估两种深度学习方法的偶然不确定性

DeepUQ: Assessing the Aleatoric Uncertainties from two Deep Learning Methods

Fermi National Accelerator Laboratory(费米国家加速器实验室);University of Chicago(芝加哥大学);Kavli Institute for Cosmological Physics(卡弗里宇宙物理研究所)

Rebecca Nevin, Aleksandra Ćiprijanović, Brian D. Nord

AI总结:

本文系统比较了Deep Ensembles和Deep Evidential Regression在0D和2D数据上的偶然不确定性量化表现,发现预测不确定性随噪声缩放但存在校准偏差,表明高噪声高维场景需进一步事后校准。

摘要:

在科学背景下,评估来自不确定性量化(UQ)深度学习方法的偶然不确定性估计的质量非常重要,因为不确定性具有物理意义,对其进行精确表征和解释至关重要。我们系统地比较了由两种UQ技术——深度集成(DE)和深度证据回归(DER)——测量的偶然不确定性。我们的方法聚焦于零维(0D)和二维(2D)数据,以探究UQ方法在不同数据维度下如何运作。我们研究了注入到输入和输出变量的不确定性,并包含了一种在输入不确定性情况下传播不确定性的方法,以便将预测的偶然不确定性与已知值进行比较。我们实验了三个噪声水平。所有模型和实验中预测的偶然不确定性均随注入的噪声水平而缩放。然而,对于一半的DE实验和几乎所有的DER实验,预测的不确定性在 $\ m{std}(σ_{\ m al})$ 上与真实不确定性存在校准偏差。对于2D输入不确定性实验和高噪声水平,这两种UQ方法预测的不确定性准确性最低。虽然这些结果不适用于更复杂的数据,但它们表明,对这些方法进行事后校准的进一步研究将是有益的,特别是在高噪声和高维设置中。

英文摘要:

Assessing the quality of aleatoric uncertainty estimates from uncertainty quantification (UQ) deep learning methods is important in scientific contexts, where uncertainty is physically meaningful and important to characterize and interpret exactly. We systematically compare aleatoric uncertainty measured by two UQ techniques, Deep Ensembles (DE) and Deep Evidential Regression (DER). Our method focuses on both zero-dimensional (0D) and two-dimensional (2D) data, to explore how the UQ methods function for different data dimensionalities. We investigate uncertainty injected on the input and output variables and include a method to propagate uncertainty in the case of input uncertainty so that we can compare the predicted aleatoric uncertainty to the known values. We experiment with three levels of noise. The aleatoric uncertainty predicted across all models and experiments scales with the injected noise level. However, the predicted uncertainty is miscalibrated to $\rm{std}(σ_{\rm al})$ with the true uncertainty for half of the DE experiments and almost all of the DER experiments. The predicted uncertainty is the least accurate for both UQ methods for the 2D input uncertainty experiment and the high-noise level. While these results do not apply to more complex data, they highlight that further research on post-facto calibration for these methods would be beneficial, particularly for high-noise and high-dimensional settings.

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。

↑