Catching Every Ripple: Enhanced Anomaly Awareness via Dynamic Concept Adaptation

2026-04-16

AI Summary

This work addresses the challenge of efficiently handling concept drift in online anomaly detection, where existing methods often rely on costly retraining or static decision boundaries. The authors propose DyMETER, a novel framework that introduces a dynamic, retraining-free adaptation mechanism. It first learns a static detector and then employs a hypernetwork to generate instance-aware parameter offsets that adapt to emerging concepts. Integrated with a lightweight evolutionary controller and a candidate window–based dynamic threshold calibration module, DyMETER enables continuous and precise anomaly detection. By jointly leveraging instance-level concept uncertainty estimation and online parameter adjustment, the method achieves superior performance across multiple real-world scenarios, demonstrating high accuracy, robustness, and computational efficiency compared to state-of-the-art approaches.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Machine Learning: Evolutionary Learning
• Search and Optimization: Evolutionary Computation

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Online anomaly detection (OAD) plays a pivotal role in real-time analytics and decision-making for evolving data streams. However, existing methods often rely on costly retraining and rigid decision boundaries, limiting their ability to adapt both effectively and efficiently to concept drift in dynamic environments. To address these challenges, we propose DyMETER, a dynamic concept adaptation framework for OAD that unifies on-the-fly parameter shifting and dynamic thresholding within a single online paradigm. DyMETER first learns a static detector on historical data to capture recurring central concepts, and then transitions to a dynamic mode to adapt to new concepts as drift occurs. Specifically, DyMETER employs a novel dynamic concept adaptation mechanism that leverages a hypernetwork to generate instance-aware parameter shifts for the static detector, thereby enabling efficient and effective adaptation without retraining or fine-tuning. To achieve robust and interpretable adaptation, DyMETER introduces a lightweight evolution controller to estimate instance-level concept uncertainty for adaptive updates. Further, DyMETER employs a dynamic threshold optimization module to adaptively recalibrates the decision boundary by maintaining a candidate window of uncertain samples, which ensures continuous alignment with evolving concepts. Extensive experiments demonstrate that DyMETER significantly outperforms existing OAD approaches across a wide spectrum of application scenarios.

Problem

• online anomaly detection
• concept drift
• dynamic adaptation
• decision boundary
• data streams

Innovation

• dynamic concept adaptation
• online anomaly detection
• hypernetwork
• concept drift
• adaptive thresholding