Hierarchical Semi-Supervised Contrastive Learning for Contamination-Resistant Anomaly Detection

TL;DR

This paper introduces HSCL, a hierarchical semi-supervised contrastive learning framework that improves anomaly detection by effectively utilizing contaminated training data without fine-tuning, achieving state-of-the-art results.

Contribution

The paper proposes a novel HSCL framework that hierarchically models multiple data relations for contamination-resistant anomaly detection, eliminating the need for fine-tuning.

Findings

HSCL achieves state-of-the-art performance in one-class classification.

HSCL effectively handles contaminated training data.

Ablation studies confirm the importance of each relation modeled.

Abstract

Anomaly detection aims at identifying deviant samples from the normal data distribution. Contrastive learning has provided a successful way to sample representation that enables effective discrimination on anomalies. However, when contaminated with unlabeled abnormal samples in training set under semi-supervised settings, current contrastive-based methods generally 1) ignore the comprehensive relation between training data, leading to suboptimal performance, and 2) require fine-tuning, resulting in low efficiency. To address the above two issues, in this paper, we propose a novel hierarchical semi-supervised contrastive learning (HSCL) framework, for contamination-resistant anomaly detection. Specifically, HSCL hierarchically regulates three complementary relations: sample-to-sample, sample-to-prototype, and normal-to-abnormal relations, enlarging the discrimination between normal and abnormal samples with a comprehensive exploration of the contaminated data. Besides, HSCL is an end-to-end learning approach that can efficiently learn discriminative representations without fine-tuning. HSCL achieves state-of-the-art performance in multiple scenarios, such as one-class classification and cross-dataset detection. Extensive ablation studies further verify the effectiveness of each considered relation. The code is available at https://github.com/GaoangW/HSCL.