Emerging applications, such as robot-assisted eldercare and object recognition, generally employ deep learning neural networks (DNNs) models and naturally require: i) handling streaming-in inference requests and ii) adapting to possible deployment scenario changes. Online model fine-tuning is widely adopted to satisfy these needs. However, fine-tuning involves significant energy consumption, making it challenging to deploy on edge devices. In this paper, we propose EdgeOL, an edge online learning framework that optimizes inference accuracy, fine-tuning execution time, and energy efficiency through both inter-tuning and intra-tuning optimizations. Experimental results show that, on average, EdgeOL reduces overall fine-tuning execution time by 82%, energy consumption by 74%, and improves average inference accuracy by 1.70% over the immediate online learning strategy.

翻译：新兴应用（如机器人辅助养老和物体识别）通常采用深度神经网络（DNNs）模型，并自然要求：i）处理流式推理请求，ii）适应可能的部署场景变化。在线模型微调被广泛采用以满足这些需求。然而，微调涉及大量能耗，使其在边缘设备上部署充满挑战。本文提出EdgeOL，一种通过调优间与调优内优化来提升推理精度、微调执行时间及能效的边缘在线学习框架。实验结果表明，与即时在线学习策略相比，EdgeOL平均减少微调执行时间82%、降低能耗74%，并提升平均推理精度1.70%。