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 2303.13845cs.CV
收藏

分布偏移下的异常检测

Anomaly Detection under Distribution Shift

Singapore Management University(新加坡管理大学)

Tri Cao, Jiawen Zhu, Guansong Pang

AI总结:

本文提出一种无监督鲁棒异常检测方法,通过最小化分布内与分布外正常样本的分布差距,应对测试数据分布偏移,在四个基准数据集上显著优于现有方法。

摘要:

异常检测(AD)是一项关键的机器学习任务,旨在从一组正常训练样本中学习模式,以识别测试数据中的异常样本。大多数现有的异常检测研究假设训练数据和测试数据来自同一数据分布,但在许多实际应用中,由于不同的自然变化,如新的光照条件、物体姿态或背景外观,测试数据可能发生较大的分布偏移,导致现有的异常检测方法在这种情况下失效。在本文中,我们考虑了分布偏移下的异常检测问题,并在四个广泛使用的异常检测和分布外(OOD)泛化数据集上建立了性能基准。我们证明,将最先进的OOD泛化方法简单适应于异常检测场景,由于缺乏标记的异常数据而无法有效工作。我们进一步提出了一种新颖的鲁棒异常检测方法,通过在训练和推理阶段以无监督方式最小化分布内和OOD正常样本之间的分布差距,来应对多样的分布偏移。我们在四个数据集上的大量实证结果表明,我们的方法在各种分布偏移的数据上显著优于最先进的异常检测方法和OOD泛化方法,同时保持了在分布内数据上的检测准确性。代码和数据可在 https://github.com/mala-lab/ADShift 获取。

英文摘要:

Anomaly detection (AD) is a crucial machine learning task that aims to learn patterns from a set of normal training samples to identify abnormal samples in test data. Most existing AD studies assume that the training and test data are drawn from the same data distribution, but the test data can have large distribution shifts arising in many real-world applications due to different natural variations such as new lighting conditions, object poses, or background appearances, rendering existing AD methods ineffective in such cases. In this paper, we consider the problem of anomaly detection under distribution shift and establish performance benchmarks on four widely-used AD and out-of-distribution (OOD) generalization datasets. We demonstrate that simple adaptation of state-of-the-art OOD generalization methods to AD settings fails to work effectively due to the lack of labeled anomaly data. We further introduce a novel robust AD approach to diverse distribution shifts by minimizing the distribution gap between in-distribution and OOD normal samples in both the training and inference stages in an unsupervised way. Our extensive empirical results on the four datasets show that our approach substantially outperforms state-of-the-art AD methods and OOD generalization methods on data with various distribution shifts, while maintaining the detection accuracy on in-distribution data. Code and data are available at https://github.com/mala-lab/ADShift.

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。

↑