NetAdapt: Platform-Aware Neural Network Adaptation for Mobile Applications
Name
1804.03230.pdf
Description
Accepted version
Size
1.44 MB
Format
Adobe PDF
Checksum (MD5)
4699c90e504d5a0f2699989b4be2e536
Author(s)
Yang, Tien-Ju
Howard, Andrew
Chen, Bo
Zhang, Xiao
Go, Alec
Sandler, Mark
Sze, Vivienne
Adam, Hartwig
Date Issued
2018
Publisher
Springer International Publishing
Citation
Yang, Tien-Ju, Howard, Andrew, Chen, Bo, Zhang, Xiao, Go, Alec et al. 2018. "NetAdapt: Platform-Aware Neural Network Adaptation for Mobile Applications."
Version
Author's final manuscript
Abstract
© Springer Nature Switzerland AG 2018. This work proposes an algorithm, called NetAdapt, that automatically adapts a pre-trained deep neural network to a mobile platform given a resource budget. While many existing algorithms simplify networks based on the number of MACs or weights, optimizing those indirect metrics may not necessarily reduce the direct metrics, such as latency and energy consumption. To solve this problem, NetAdapt incorporates direct metrics into its adaptation algorithm. These direct metrics are evaluated using empirical measurements, so that detailed knowledge of the platform and toolchain is not required. NetAdapt automatically and progressively simplifies a pre-trained network until the resource budget is met while maximizing the accuracy. Experiment results show that NetAdapt achieves better accuracy versus latency trade-offs on both mobile CPU and mobile GPU, compared with the state-of-the-art automated network simplification algorithms. For image classification on the ImageNet dataset, NetAdapt achieves up to a 1.7 × speedup in measured inference latency with equal or higher accuracy on MobileNets (V1&V2).
MIT Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1007/978-3-030-01249-6_18
