通过融合特征密度提升不确定性估计，显著增强模型对异常样本的识别能力
Uncertainty Estimation by Density Aware Evidential Deep Learning
- 将测试样本在特征空间中的密度信息与传统证据深度学习结合
- 在多个数据集上实现最优的异常检测与分类准确率
- 特别适合需要可靠不确定度评估的医疗、自动驾驶等安全关键场景
证据深度学习（EDL）在不确定性估计方面表现优异，但在分布外（OOD）检测和分类任务中仍有提升空间。当前EDL在OOD检测上的局限源于其无法反映测试样本与训练数据之间的距离，而分类性能受限则来自浓度参数的参数化方式。为此，本文提出一种新方法——密度感知证据深度学习（DAEDL）。DAEDL在预测阶段将测试样本的特征空间密度与EDL输出相结合，并采用新型参数化方式解决传统参数化的缺陷。理论分析表明，DAEDL具备多项优良性质。实验结果显示，DAEDL在多种与不确定性估计和分类相关的下游任务中均达到领先水平。
原文摘要 · Abstract (English)
Evidential deep learning (EDL) has shown remarkable success in uncertainty estimation. However, there is still room for improvement, particularly in out-of-distribution (OOD) detection and classification tasks. The limited OOD detection performance of EDL arises from its inability to reflect the distance between the testing example and training data when quantifying uncertainty, while its limited classification performance stems from its parameterization of the concentration parameters. To address these limitations, we propose a novel method called Density Aware Evidential Deep Learning (DAEDL). DAEDL integrates the feature space density of the testing example with the output of EDL during the prediction stage, while using a novel parameterization that resolves the issues in the conventional parameterization. We prove that DAEDL enjoys a number of favorable theoretical properties. DAEDL demonstrates state-of-the-art performance across diverse downstream tasks related to uncertainty estimation and classification
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。