Link-Aware Energy-Frugal Continual Learning for Fault Detection in IoT Networks
This addresses energy-efficient fault detection for resource-constrained IoT devices, representing an incremental improvement in adaptive continual learning.
The paper tackles the problem of maintaining inference accuracy for lightweight ML models in IoT networks under non-stationary environments and energy constraints by introducing an event-driven communication framework for continual learning, achieving up to a 42.8% improvement in recall compared to baseline methods.
The use of lightweight machine learning (ML) models in internet of things (IoT) networks enables resource constrained IoT devices to perform on-device inference for several critical applications. However, the inference accuracy deteriorates due to the non-stationarity in the IoT environment and limited initial training data. To counteract this, the deployed models can be updated occasionally with new observed data samples. However, this approach consumes additional energy, which is undesirable for energy constrained IoT devices. This letter introduces an event-driven communication framework that strategically integrates continual learning (CL) in IoT networks for energy-efficient fault detection. Our framework enables the IoT device and the edge server (ES) to collaboratively update the lightweight ML model by adapting to the wireless link conditions for communication and the available energy budget. Evaluation on real-world datasets show that the proposed approach can outperform both periodic sampling and non-adaptive CL in terms of inference recall; our proposed approach achieves up to a 42.8% improvement, even under tight energy and bandwidth constraint.