AdaEvo: Edge-Assisted Continuous and Timely DNN Model Evolution for Mobile Devices

Mobile video applications today have attracted significant attention. Deep learning model (e.g. deep neural network, DNN) compression is widely used to enable on-device inference for facilitating robust and private mobile video applications. The compressed DNN, however, is vulnerable to the agnostic...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Wang, Lehao, Yu, Zhiwen, Yu, Haoyi, Liu, Sicong, Xie, Yaxiong, Guo, Bin, Liu, Yunxin
Format Paper
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 25.10.2023
Subjects
Accuracy
Artificial neural networks
Data collection
Drift
Edge computing
Evolution
Machine learning
Mobile computing
Servers
Streaming media
Video compression
Online AccessGet full text

Cover

More Information
Summary:Mobile video applications today have attracted significant attention. Deep learning model (e.g. deep neural network, DNN) compression is widely used to enable on-device inference for facilitating robust and private mobile video applications. The compressed DNN, however, is vulnerable to the agnostic data drift of the live video captured from the dynamically changing mobile scenarios. To combat the data drift, mobile ends rely on edge servers to continuously evolve and re-compress the DNN with freshly collected data. We design a framework, AdaEvo, that efficiently supports the resource-limited edge server handling mobile DNN evolution tasks from multiple mobile ends. The key goal of AdaEvo is to maximize the average quality of experience (QoE), e.g. the proportion of high-quality DNN service time to the entire life cycle, for all mobile ends. Specifically, it estimates the DNN accuracy drops at the mobile end without labels and performs a dedicated video frame sampling strategy to control the size of retraining data. In addition, it balances the limited computing and memory resources on the edge server and the competition between asynchronous tasks initiated by different mobile users. With an extensive evaluation of real-world videos from mobile scenarios and across four diverse mobile tasks, experimental results show that AdaEvo enables up to 34% accuracy improvement and 32% average QoE improvement.
ISSN:2331-8422