[论文解读] Explainable Deep Few-shot Anomaly Detection with Deviation Networks

Guansong Pang, Choubo Ding+2 more|arXiv (Cornell University)|Aug 1, 2021

Anomaly Detection Techniques and Applications参考文献 68被引用 43

一句话总结

提出 DevNet，一种基于偏差的端到端深度框架用于小样本异常检测，直接学习以高斯先验和 MIL 基于偏差损失引导的异常分数，从而实现开放集检测和可解释性分数。

ABSTRACT

Existing anomaly detection paradigms overwhelmingly focus on training detection models using exclusively normal data or unlabeled data (mostly normal samples). One notorious issue with these approaches is that they are weak in discriminating anomalies from normal samples due to the lack of the knowledge about the anomalies. Here, we study the problem of few-shot anomaly detection, in which we aim at using a few labeled anomaly examples to train sample-efficient discriminative detection models. To address this problem, we introduce a novel weakly-supervised anomaly detection framework to train detection models without assuming the examples illustrating all possible classes of anomaly. Specifically, the proposed approach learns discriminative normality (regularity) by leveraging the labeled anomalies and a prior probability to enforce expressive representations of normality and unbounded deviated representations of abnormality. This is achieved by an end-to-end optimization of anomaly scores with a neural deviation learning, in which the anomaly scores of normal samples are imposed to approximate scalar scores drawn from the prior while that of anomaly examples is enforced to have statistically significant deviations from these sampled scores in the upper tail. Furthermore, our model is optimized to learn fine-grained normality and abnormality by top-K multiple-instance-learning-based feature subspace deviation learning, allowing more generalized representations. Comprehensive experiments on nine real-world image anomaly detection benchmarks show that our model is substantially more sample-efficient and robust, and performs significantly better than state-of-the-art competing methods in both closed-set and open-set settings. Our model can also offer explanation capability as a result of its prior-driven anomaly score learning. Code and datasets are available at: https://git.io/DevNet.

研究动机与目标

• 激发并解决仅有少量标注异常可用的少样本异常检测问题。
• 学习一个辨别性强、鲁棒的正常性模型，同时处理未见异常。
• 提供端到端的异常分数学习，而不是间接的特征学习。
• 通过与学习分数相关的梯度定位实现异常解释。

提出的方法

• DevNet 直接学习一个将输入映射到标量异常分数的异常评分函数。
• 使用基于高斯先验的参考分数来引导端到端的异常分数学习。
• 基于 MIL 的 Top-K 偏差损失确保异常的异常分数在上尾偏离先验，而正常分数接近先验。
• 端到端网络结合细粒度特征学习器和一个简单线性异常分数头。
• 通过梯度反向传播实现异常解释，以定位贡献的特征。

实验结果

研究问题

• RQ1 少量标注异常是否能引导端到端学习异常分数，用于已见和未见的异常类别？
• RQ2 利用高斯先验和 MIL 基偏差损失是否提升 FSAD 的样本效率和泛化能力？
• RQ3 模型是否能提供与学习分数相关的真实的异常定位/解释？
• RQ4 该方法对正常数据中的异常污染是否鲁棒，对不同骨干网络和数据集是否有效？

主要发现

• DevNet 在九个真实世界图像异常基准上实现了改进的样本效率和鲁棒性，优于最新方法。
• 在开放集设置中，该框架对未见异常类别具有更好的泛化能力。
• 由于先验驱动学习，异常分数具有可解释性，并能通过梯度对输入区域进行定位解释。
• Top-K MIL 偏差学习通过聚焦最具信息量的异常补丁提升学习，减少误检。
• DevNet 提供准确的异常定位，支持对检测到的异常的解释。