The need to train DNN models on end-user devices (e.g., smartphones) is increasing with the need to improve data privacy and reduce communication overheads. Unlike datacenter servers with powerful CPUs and GPUs, modern smartphones consist of a diverse collection of specialized cores following a system-on-a-chip (SoC) architecture that together perform a variety of tasks. We observe that training DNNs on a smartphone SoC without carefully considering its resource constraints can not only lead to suboptimal training performance but significantly affect user experience as well. In this paper, we present Swan, a neural engine to optimize DNN training on smartphone SoCs without hurting user experience. Extensive large-scale evaluations show that Swan can improve performance by 1.2 - 23.3x over the state-of-the-art.
@article{arxiv.2206.04687,
title = {Swan: A Neural Engine for Efficient DNN Training on Smartphone SoCs},
author = {Sanjay Sri Vallabh Singapuram and Fan Lai and Chuheng Hu and Mosharaf Chowdhury},
journal= {arXiv preprint arXiv:2206.04687},
year = {2022}
}