MECAD: A multi-expert architecture for continual anomaly detection

TL;DR

MECAD introduces a multi-expert architecture for continual anomaly detection that dynamically assigns experts to object classes, efficiently preserves knowledge, and demonstrates high performance on diverse industrial datasets.

Contribution

It presents a novel multi-expert system with dynamic expert assignment and memory management for incremental anomaly detection without full retraining.

Findings

Achieves an average AUROC of 0.8259 on MVTec AD dataset.

Reduces knowledge degradation compared to single-expert methods.

Balances efficiency, knowledge retention, and adaptability.

Abstract

In this paper we propose MECAD, a novel approach for continual anomaly detection using a multi-expert architecture. Our system dynamically assigns experts to object classes based on feature similarity and employs efficient memory management to preserve the knowledge of previously seen classes. By leveraging an optimized coreset selection and a specialized replay buffer mechanism, we enable incremental learning without requiring full model retraining. Our experimental evaluation on the MVTec AD dataset demonstrates that the optimal 5-expert configuration achieves an average AUROC of 0.8259 across 15 diverse object categories while significantly reducing knowledge degradation compared to single-expert approaches. This framework balances computational efficiency, specialized knowledge retention, and adaptability, making it well-suited for industrial environments with evolving product types.