# ARCADe: A Rapid Continual Anomaly Detector

arXiv: https://arxiv.org/abs/2008.04042 · HTML: https://mlsanity.com/paper/2008.04042/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Aug 2020.

## Built on by (most cited first, 7 of 7)

- [Unveiling the Anomalies in an Ever-Changing World: A Benchmark for Pixel-Level Anomaly Detection in Continual Learning](https://mlsanity.com/paper/2403.15463/) (2024)
- [Towards Data-Free Domain Generalization](https://mlsanity.com/paper/2110.04545/) (2021)
- [Continual Learning Approaches for Anomaly Detection](https://mlsanity.com/paper/2212.11192/) (2022)
- [Lifelong Continual Learning for Anomaly Detection: New Challenges, Perspectives, and Insights](https://mlsanity.com/paper/2303.07557/) (2023)
- [Deep evolving semi-supervised anomaly detection](https://mlsanity.com/paper/2412.00860/) (2024)
- [One-Class Domain Adaptation via Meta-Learning](https://mlsanity.com/paper/2501.13052/) (2025)
- [TaskFusion: Continual Anomaly Detection for Heterogeneous Tabular Data](https://mlsanity.com/paper/2606.11844/) (2026)

## Abstract (arXiv)

Although continual learning and anomaly detection have separately been well-studied in previous works, their intersection remains rather unexplored. The present work addresses a learning scenario where a model has to incrementally learn a sequence of anomaly detection tasks, i.e. tasks from which only examples from the normal (majority) class are available for training. We define this novel learning problem of continual anomaly detection (CAD) and formulate it as a meta-learning problem. Moreover, we propose A Rapid Continual Anomaly Detector (ARCADe), an approach to train neural networks to be robust against the major challenges of this new learning problem, namely catastrophic forgetting and overfitting to the majority class. The results of our experiments on three datasets show that, in the CAD problem setting, ARCADe substantially outperforms baselines from the continual learning and anomaly detection literature. Finally, we provide deeper insights into the learning strategy yielded by the proposed meta-learning algorithm.