ONER: Online Experience Replay for Incremental Anomaly Detection

2024-12-05

arXiv.org

AI Summary

To address catastrophic forgetting in incremental anomaly detection for dynamic industrial scenarios—where learning new anomaly classes erodes prior knowledge and induces feature conflicts between old and new classes—this paper proposes an end-to-end online incremental learning framework. Methodologically, it introduces (1) the first online experience replay mechanism tailored for anomaly detection; (2) a coupled architecture integrating decomposable attention prompting with pixel- and image-level multi-granularity semantic prototypes to jointly suppress forgetting via parameter update regularization and feature space optimization; and (3) learnable prompt assembly and attention-conditioned prompt generation to enhance generalization across tasks. Evaluated on standard incremental anomaly detection benchmarks, the framework achieves state-of-the-art performance, significantly reducing forgetting rates while incurring minimal adaptation overhead for new classes—demonstrating both computational efficiency and training stability.

Technology Category

• Machine Learning: Online Learning & Bandits
• Search and Optimization: Learning to Search
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Incremental anomaly detection sequentially recognizes abnormal regions in novel categories for dynamic industrial scenarios. This remains highly challenging due to knowledge overwriting and feature conflicts, leading to catastrophic forgetting. In this work, we propose ONER, an end-to-end ONline Experience Replay method, which efficiently mitigates catastrophic forgetting while adapting to new tasks with minimal cost. Specifically, our framework utilizes two types of experiences from past tasks: decomposed prompts and semantic prototypes, addressing both model parameter updates and feature optimization. The decomposed prompts consist of learnable components that assemble to produce attention-conditioned prompts. These prompts reuse previously learned knowledge, enabling model to learn novel tasks effectively. The semantic prototypes operate at both pixel and image levels, performing regularization in the latent feature space to prevent forgetting across various tasks. Extensive experiments demonstrate that our method achieves state-of-the-art performance in incremental anomaly detection with significantly reduced forgetting, as well as efficiently adapting to new categories with minimal costs. These results confirm the efficiency and stability of ONER, making it a powerful solution for real-world applications.

Problem

• Incremental Anomaly Detection
• Dynamic Industrial Scenarios
• Knowledge Consolidation

Innovation

• Online Experience Replay
• Incremental Anomaly Detection
• Decomposed Prompts and Semantic Prototypes