Understanding the limitations of self-supervised learning for tabular anomaly detection
While self-supervised learning has improved anomaly detection in computer vision and natural language processing, it is unclear whether tabular data can benefit from it. This paper explores the limitations of self-supervision for tabular anomaly detection. We conduct several experiments spanning various pretext tasks on 26 benchmark datasets to understand why this is the case. Our results confirm representations derived from self-supervision do not improve tabular anomaly detection performance compared to using the raw representations of the data. We show this is due to neural networks introducing irrelevant features, which reduces the effectiveness of anomaly detectors. However, we demonstrate that using a subspace of the neural network's representation can recover performance.
Code (0)
등록된 구현이 없습니다.
Tasks
Anomaly DetectionSelf-Supervised LearningSimilar Papers 제목 키워드 기반
SORTAD: Self-Supervised Optimized Random Transformations for Anomaly Detection in Tabular Data
We consider a self-supervised approach to anomaly detection in tabular data. Random transformations are applied to the data, and then each transformation is identified based on its output. These predicted transformations…
Anomaly DetectionSelf-supervise, Refine, Repeat: Improving Unsupervised Anomaly Detection
Anomaly detection (AD), separating anomalies from normal data, has many applications across domains, from security to healthcare. While most previous works were shown to be effective for cases with fully or partially lab…
Anomaly DetectionClassificationOne-Class ClassificationOne-class classifier+2Enhancing Tabular Anomaly Detection via Pseudo-Label-Guided Generation
Identifying anomalous instances in tabular data is essential for improving data reliability and maintaining system stability. Due to the scarcity of ground-truth anomaly labels, existing methods mainly rely on unsupervis…
Unsupervised Anomaly DetectionContrastive LearningICLAD: In-Context Learning for Unified Tabular Anomaly Detection Across Supervision Regimes
Anomaly detection on tabular data is commonly studied under three supervision regimes, including one-class settings that assume access to anomaly-free training samples, fully unsupervised settings with unlabeled and pote…
Anomaly DetectionRGLD: Randomized Global-Local Density Estimation for Tabular Anomaly Detection
Unsupervised tabular anomaly detection requires methods that are accurate, robust across heterogeneous datasets, and computationally efficient. Classical statistical detectors are often efficient, but they usually rely o…
Density EstimationAnomaly Detection