Abstract
<jats:p>This research investigates the efficacy of various continual learning techniques—Elastic WeightConsolidation (EWC), Experience Replay, and Knowledge Distillation—within the framework of EdgeAI for real-time applications. Continual learning, crucial for adapting AI models to new data whilepreserving previously acquired knowledge, presents unique challenges when deployed on resource-constrained edge devices. This study evaluates these techniques based on key performance metricsincluding task accuracy, old task accuracy, latency, resource usage, and adaptability. The findings revealthat Experience Replay excels in maintaining high task accuracy and adaptability, albeit with increasedresource demands. EWC provides a balanced approach with moderate performance and resource usagebut shows slightly lower adaptability. Knowledge Distillation offers an efficient solution with goodperformance and minimal computational overhead, making it suitable for edge environments with strictresource constraints. These insights guide the selection of continual learning methods tailored to thespecific needs of real-time Edge AI applications</jats:p>