Online Continual Learning for Anomaly Detection in IoT under Data Distribution Shifts
This work provides an incremental solution for maintaining accurate anomaly detection models on resource-constrained IoT devices under changing data distributions.
This paper addresses anomaly detection in IoT devices operating in non-stationary environments where data distribution shifts make on-device models obsolete. The proposed OCLADS framework uses intelligent sample selection at the device and distribution-shift detection at the edge server to timely update the anomaly detection model, achieving high inference accuracy with significantly fewer model updates compared to baselines.
In this work, we present OCLADS, a novel communication framework with continual learning (CL) for Internet of Things (IoT) anomaly detection (AD) when operating in non-stationary environments. As the statistical properties of the observed data change with time, the on-device inference model becomes obsolete, which necessitates strategic model updating. OCLADS keeps track of data distribution shifts to timely update the on-device IoT AD model. To do so, OCLADS introduces two mechanisms during the interaction between the resource-constrained IoT device and an edge server (ES): i) an intelligent sample selection mechanism at the device for data transmission, and ii) a distribution-shift detection mechanism at the ES for model updating. Experimental results with TinyML demonstrate that our proposed framework achieves high inference accuracy while realizing a significantly smaller number of model updates compared to the baseline schemes.