通过归一化离群分布自适应的长尾分布外检测
Long-Tailed Out-of-Distribution Detection via Normalized Outlier Distribution Adaptation
AI总结：
针对长尾识别中伪 OOD 样本与真实 OOD 样本存在分布偏移的问题，本文提出 AdaptOD 方法，通过动态离群分布自适应和双归一化能量损失学习更可靠的 OOD 检测器，在多个基准上取得最优性能。
摘要：
分布外（OOD）检测的一个关键挑战是训练期间缺乏真实 OOD 样本。解决该问题的一种原则性方法是使用外部数据集中的样本作为离群样本（即伪 OOD 样本）来训练 OOD 检测器。然而，我们通过实验发现，离群样本通常与真实 OOD 样本存在分布偏移，尤其是在长尾识别（LTR）场景中，ID 类别严重不平衡，即真实 OOD 样本与离群样本中的头部和尾部 ID 类别表现出非常不同的概率分布。在本工作中，我们提出了一种新方法，即归一化离群分布自适应（AdaptOD），以解决这一分布偏移问题。其关键组成部分之一是动态离群分布自适应，它利用推理过程中预测 OOD 样本中的 OOD 知识，有效地将基于离群样本的普通离群分布自适应到真实 OOD 分布。此外，为了在长尾 ID 数据上获得更可靠的预测 OOD 样本集合，AdaptOD 中引入了一种新颖的双归一化能量损失，它利用类别级和样本级归一化能量，在不平衡的 ID 样本上强制实现更平衡的预测能量。这有助于避免对头部样本的偏置，并在训练期间学习比现有能量损失显著更好的普通离群分布。它还消除了手动调节能量损失中敏感边际超参数的需要。在三个流行的 LTR 中 OOD 检测基准上的实验结果表明，AdaptOD 优于最先进的方法。代码可在 https://github.com/mala-lab/AdaptOD 获取。
英文摘要：
One key challenge in Out-of-Distribution (OOD) detection is the absence of ground-truth OOD samples during training. One principled approach to address this issue is to use samples from external datasets as outliers (i.e., pseudo OOD samples) to train OOD detectors. However, we find empirically that the outlier samples often present a distribution shift compared to the true OOD samples, especially in Long-Tailed Recognition (LTR) scenarios, where ID classes are heavily imbalanced, \ie, the true OOD samples exhibit very different probability distribution to the head and tailed ID classes from the outliers. In this work, we propose a novel approach, namely normalized outlier distribution adaptation (AdaptOD), to tackle this distribution shift problem. One of its key components is dynamic outlier distribution adaptation that effectively adapts a vanilla outlier distribution based on the outlier samples to the true OOD distribution by utilizing the OOD knowledge in the predicted OOD samples during inference. Further, to obtain a more reliable set of predicted OOD samples on long-tailed ID data, a novel dual-normalized energy loss is introduced in AdaptOD, which leverages class- and sample-wise normalized energy to enforce a more balanced prediction energy on imbalanced ID samples. This helps avoid bias toward the head samples and learn a substantially better vanilla outlier distribution than existing energy losses during training. It also eliminates the need of manually tuning the sensitive margin hyperparameters in energy losses. Empirical results on three popular benchmarks for OOD detection in LTR show the superior performance of AdaptOD over state-of-the-art methods. Code is available at https://github.com/mala-lab/AdaptOD.
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。
