EdgeOL: Efficient in-situ Online Learning on Edge Devices
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866918021738528768 |
|---|---|
| author | Li, Sheng Yuan, Geng Dai, Yue Wang, Tianyu Wu, Yawen Jones, Alex K. Hu, Jingtong Tony Geng Wang, Yanzhi Yuan, Bo Ding, Yufei Tang, Xulong |
| author_facet | Li, Sheng Yuan, Geng Dai, Yue Wang, Tianyu Wu, Yawen Jones, Alex K. Hu, Jingtong Tony Geng Wang, Yanzhi Yuan, Bo Ding, Yufei Tang, Xulong |
| contents | Emerging applications, such as robot-assisted eldercare and object recognition, generally employ deep learning neural networks (DNNs) 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, an inappropriate fine-tuning scheme could involve 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 64%, energy consumption by 52%, and improves average inference accuracy by 1.75% over the immediate online learning strategy |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_16694 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | EdgeOL: Efficient in-situ Online Learning on Edge Devices Li, Sheng Yuan, Geng Dai, Yue Wang, Tianyu Wu, Yawen Jones, Alex K. Hu, Jingtong Tony Geng Wang, Yanzhi Yuan, Bo Ding, Yufei Tang, Xulong Machine Learning Computer Vision and Pattern Recognition Distributed, Parallel, and Cluster Computing Emerging applications, such as robot-assisted eldercare and object recognition, generally employ deep learning neural networks (DNNs) 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, an inappropriate fine-tuning scheme could involve 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 64%, energy consumption by 52%, and improves average inference accuracy by 1.75% over the immediate online learning strategy |
| title | EdgeOL: Efficient in-situ Online Learning on Edge Devices |
| topic | Machine Learning Computer Vision and Pattern Recognition Distributed, Parallel, and Cluster Computing |
| url | https://arxiv.org/abs/2401.16694 |