Neural networks are powerful tools in various applications, and quantifying their uncertainty is crucial for reliable decision-making. In the deep learning field, the uncertainties are usually categorized into aleatoric (data) and epistemic (model) uncertainty. In this paper, we point out that the existing popular variance attenuation method highly overestimates aleatoric uncertainty. To address this issue, we propose a new estimation method by actively de-noising the observed data \footnote{Source code available at \url{https://github.com/wz16/DVA}.}. By conducting a broad range of experiments, we demonstrate that our proposed approach provides a much closer approximation to the actual data uncertainty than the standard method.

翻译：神经网络在各种应用中都是强大的工具，量化其不确定性对于可靠决策至关重要。在深度学习领域，不确定性通常分为偶然不确定性（数据不确定性）和认知不确定性（模型不确定性）。在本文中，我们指出现有的流行方差衰减方法会严重高估偶然不确定性。为解决这一问题，我们提出一种新的估计方法，通过对观测数据进行主动去噪来实现\footnote{源代码可在\url{https://github.com/wz16/DVA}获取。}。通过开展广泛的实验，我们证明与标准方法相比，我们提出的方法能够更接近实际的数据不确定性近似。