OCSVM-Guided Representation Learning for Unsupervised Anomaly Detection

2025-07-25

AI Summary

Unsupervised anomaly detection (UAD) faces two key challenges: reconstruction-based methods tend to overfit anomalies, while disentangled representation learning coupled with density estimators lacks discriminative power in feature space. This paper proposes OCSVM-Align—a framework that end-to-end couples deep representation learning with analytically tractable one-class support vector machines (OCSVM), directly aligning the latent-space feature distribution with the optimal OCSVM hypersphere boundary—eliminating surrogate objectives and kernel approximations. A differentiable OCSVM loss enables joint optimization of feature extraction and anomaly discrimination. Evaluated on MNIST-C and brain MRI datasets, OCSVM-Align achieves significant improvements in pixel-level anomaly detection accuracy, demonstrating robustness to small-sized and low-contrast lesions. It establishes new state-of-the-art performance, underscoring strong potential for clinical deployment.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Machine Learning: Unsupervised & Self-Supervised Learning
• Computer Vision: Representation Learning for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Unsupervised anomaly detection (UAD) aims to detect anomalies without labeled data, a necessity in many machine learning applications where anomalous samples are rare or not available. Most state-of-the-art methods fall into two categories: reconstruction-based approaches, which often reconstruct anomalies too well, and decoupled representation learning with density estimators, which can suffer from suboptimal feature spaces. While some recent methods attempt to couple feature learning and anomaly detection, they often rely on surrogate objectives, restrict kernel choices, or introduce approximations that limit their expressiveness and robustness. To address this challenge, we propose a novel method that tightly couples representation learning with an analytically solvable one-class SVM (OCSVM), through a custom loss formulation that directly aligns latent features with the OCSVM decision boundary. The model is evaluated on two tasks: a new benchmark based on MNIST-C, and a challenging brain MRI subtle lesion detection task. Unlike most methods that focus on large, hyperintense lesions at the image level, our approach succeeds to target small, non-hyperintense lesions, while we evaluate voxel-wise metrics, addressing a more clinically relevant scenario. Both experiments evaluate a form of robustness to domain shifts, including corruption types in MNIST-C and scanner/age variations in MRI. Results demonstrate performance and robustness of our proposed mode,highlighting its potential for general UAD and real-world medical imaging applications. The source code is available at https://github.com/Nicolas-Pinon/uad_ocsvm_guided_repr_learning

Problem

• Improves unsupervised anomaly detection without labeled data
• Addresses suboptimal feature spaces in decoupled representation learning
• Targets small non-hyperintense lesions in medical imaging

Innovation

• OCSVM-guided representation learning for UAD
• Custom loss aligns features with OCSVM boundary
• Targets subtle lesions with voxel-wise metrics