Filter or Compensate: Towards Invariant Representation from Distribution Shift for Anomaly Detection

TL;DR

This paper introduces FiCo, a novel method for anomaly detection under distribution shift, which compensates for distribution-specific information and filters abnormal data to learn invariant normality, outperforming existing methods.

Contribution

FiCo addresses distribution shift in anomaly detection by combining distribution-specific compensation and invariant filtering, improving robustness and accuracy.

Findings

FiCo outperforms state-of-the-art methods on three AD benchmarks.

FiCo achieves better results than RD-based methods in ID scenarios.

Extensive experiments validate FiCo's effectiveness and robustness.

Abstract

Recent Anomaly Detection (AD) methods have achieved great success with In-Distribution (ID) data. However, real-world data often exhibits distribution shift, causing huge performance decay on traditional AD methods. From this perspective, few previous work has explored AD with distribution shift, and the distribution-invariant normality learning has been proposed based on the Reverse Distillation (RD) framework. However, we observe the misalignment issue between the teacher and the student network that causes detection failure, thereby propose FiCo, Filter or Compensate, to address the distribution shift issue in AD. FiCo firstly compensates the distribution-specific information to reduce the misalignment between the teacher and student network via the Distribution-Specific Compensation (DiSCo) module, and secondly filters all abnormal information to capture distribution-invariant normality with the Distribution-Invariant Filter (DiIFi) module. Extensive experiments on three different AD benchmarks demonstrate the effectiveness of FiCo, which outperforms all existing state-of-the-art (SOTA) methods, and even achieves better results on the ID scenario compared with RD-based methods. Our code is available at https://github.com/znchen666/FiCo.