Contrastive Representation Modeling for Anomaly Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lunardi, Willian T., Banabila, Abdulrahman, Herzalla, Dania, Andreoni, Martin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910005128593408
author Lunardi, Willian T.
Banabila, Abdulrahman
Herzalla, Dania
Andreoni, Martin
author_facet Lunardi, Willian T.
Banabila, Abdulrahman
Herzalla, Dania
Andreoni, Martin
contents Distance-based anomaly detection methods rely on compact in-distribution (ID) embeddings that are well separated from anomalies. However, conventional contrastive learning strategies often struggle to achieve this balance, either promoting excessive variance among inliers or failing to preserve the diversity of outliers. We begin by analyzing the challenges of representation learning for anomaly detection and identify three essential properties for the pretext task: (1) compact clustering of inliers, (2) strong separation between inliers and anomalies, and (3) preservation of diversity among synthetic outliers. Building on this, we propose a structured contrastive objective that redefines positive and negative relationships during training, promoting these properties without requiring explicit anomaly labels. We extend this framework with a patch-based learning and evaluation strategy specifically designed to improve the detection of localized anomalies in industrial settings. Our approach demonstrates significantly faster convergence and improved performance compared to standard contrastive methods. It matches or surpasses anomaly detection methods on both semantic and industrial benchmarks, including methods that rely on discriminative training or explicit anomaly labels.
format Preprint
id arxiv_https___arxiv_org_abs_2501_05130
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Contrastive Representation Modeling for Anomaly Detection
Lunardi, Willian T.
Banabila, Abdulrahman
Herzalla, Dania
Andreoni, Martin
Machine Learning
I.2.6; I.5.1
Distance-based anomaly detection methods rely on compact in-distribution (ID) embeddings that are well separated from anomalies. However, conventional contrastive learning strategies often struggle to achieve this balance, either promoting excessive variance among inliers or failing to preserve the diversity of outliers. We begin by analyzing the challenges of representation learning for anomaly detection and identify three essential properties for the pretext task: (1) compact clustering of inliers, (2) strong separation between inliers and anomalies, and (3) preservation of diversity among synthetic outliers. Building on this, we propose a structured contrastive objective that redefines positive and negative relationships during training, promoting these properties without requiring explicit anomaly labels. We extend this framework with a patch-based learning and evaluation strategy specifically designed to improve the detection of localized anomalies in industrial settings. Our approach demonstrates significantly faster convergence and improved performance compared to standard contrastive methods. It matches or surpasses anomaly detection methods on both semantic and industrial benchmarks, including methods that rely on discriminative training or explicit anomaly labels.
title Contrastive Representation Modeling for Anomaly Detection
topic Machine Learning
I.2.6; I.5.1
url https://arxiv.org/abs/2501.05130