Robustness to Spurious Correlations Improves Semantic Out-of-Distribution Detection
Robustness to Spurious Correlations Improves Semantic Out-of-Distribution Detection
复制标题
DOI：
10.48550/arxiv.2302.04132
复制
发表时间：
2023-02
期刊：
影响因子：
--
通讯作者：
Lily H. Zhang;R. Ranganath
中科院分区：
文献类型：
--
作者：
Lily H. Zhang;R. Ranganath
利用预测模型的输出或特征表示的方法已经成为图像输入的分布外（ood）检测的有前途的方法。然而，这些方法难以检测与分布内输入共享滋扰值（例如，背景）的良好输入。共享干扰分布外（sn-ood）输入的检测在现实世界的应用中特别相关，因为异常和分布内输入往往在部署期间在相同的设置中被捕获。在这项工作中，我们提供了一个可能的解释sn-ood检测失败，并提出滋扰意识ood检测来解决这些问题。滋扰感知的ood检测替代了通过经验风险最小化（Empirical Risk Minimization，简称ERI）和交叉熵损失训练的分类器，其中1。是在一个分布下训练的，其中滋扰标签关系被打破，2。产生独立于该分布下的干扰的表示，无论是边缘的还是以标签为条件的。我们可以训练一个分类器来实现这些目标，使用滋扰随机蒸馏（Nuisance-Randomized Distillation，简称NDID），这是一种为虚假相关下的良好泛化而开发的算法。基于输出和特征的滋扰感知ood检测性能大大优于其原来的同行，成功，即使在域泛化算法的基础上检测未能提高性能。
Methods which utilize the outputs or feature representations of predictive models have emerged as promising approaches for out-of-distribution (ood) detection of image inputs. However, these methods struggle to detect ood inputs that share nuisance values (e.g. background) with in-distribution inputs. The detection of shared-nuisance out-of-distribution (sn-ood) inputs is particularly relevant in real-world applications, as anomalies and in-distribution inputs tend to be captured in the same settings during deployment. In this work, we provide a possible explanation for sn-ood detection failures and propose nuisance-aware ood detection to address them. Nuisance-aware ood detection substitutes a classifier trained via Empirical Risk Minimization (erm) and cross-entropy loss with one that 1. is trained under a distribution where the nuisance-label relationship is broken and 2. yields representations that are independent of the nuisance under this distribution, both marginally and conditioned on the label. We can train a classifier to achieve these objectives using Nuisance-Randomized Distillation (NURD), an algorithm developed for ood generalization under spurious correlations. Output- and feature-based nuisance-aware ood detection perform substantially better than their original counterparts, succeeding even when detection based on domain generalization algorithms fails to improve performance.