arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2403.15905cs.LGcs.CV
收藏

面向资源受限设备的低能耗自适应个性化方法

Towards Low-Energy Adaptive Personalization for Resource-Constrained Devices

Imperial College London(伦敦帝国学院);University of York(约克大学)

Yushan Huang, Josh Millar, Yuxuan Long, Yuchen Zhao, Hamed Haddadi

AI总结:

本文提出目标块微调(TBFT)框架,将数据漂移分为输入、特征和输出三级并分别微调模型前、中、后块,在资源受限设备上以更低能耗实现更优个性化性能。

摘要:

机器学习(ML)模型的个性化以应对数据漂移是物联网(IoT)应用背景下的一个重大挑战。目前,大多数方法侧重于微调完整的基础模型或其最后几层以适应新数据,同时往往忽略能耗成本。然而,数据漂移存在多种类型,在某些场景下,微调完整的基础模型或最后几层可能无法获得最优性能。我们提出目标块微调(Target Block Fine-Tuning,TBFT),这是一种为资源受限设备设计的低能耗自适应个性化框架。我们将数据漂移和个性化划分为三种类型:输入级、特征级和输出级。针对每种类型,我们微调模型的不同块,以在降低能耗成本的同时实现最优性能。具体而言,输入级、特征级和输出级分别对应微调模型的前部、中部和后部块。我们在ResNet模型、三个数据集、三种不同训练规模以及Raspberry Pi上对TBFT进行了评估。与逐块微调并平均其性能提升的Block Avg相比,TBFT的模型准确率平均提高了15.30%,同时与完整微调相比平均节省了41.57%的能耗。

英文摘要:

The personalization of machine learning (ML) models to address data drift is a significant challenge in the context of Internet of Things (IoT) applications. Presently, most approaches focus on fine-tuning either the full base model or its last few layers to adapt to new data, while often neglecting energy costs. However, various types of data drift exist, and fine-tuning the full base model or the last few layers may not result in optimal performance in certain scenarios. We propose Target Block Fine-Tuning (TBFT), a low-energy adaptive personalization framework designed for resource-constrained devices. We categorize data drift and personalization into three types: input-level, feature-level, and output-level. For each type, we fine-tune different blocks of the model to achieve optimal performance with reduced energy costs. Specifically, input-, feature-, and output-level correspond to fine-tuning the front, middle, and rear blocks of the model. We evaluate TBFT on a ResNet model, three datasets, three different training sizes, and a Raspberry Pi. Compared with the $Block Avg$, where each block is fine-tuned individually and their performance improvements are averaged, TBFT exhibits an improvement in model accuracy by an average of 15.30% whilst saving 41.57% energy consumption on average compared with full fine-tuning.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑