AMC: AutoML for Model Compression and Acceleration on Mobile Devices
Name
1802.03494.pdf
Description
Accepted version
Size
792.64 KB
Format
Adobe PDF
Checksum (MD5)
bd988a64eebb1eb4e5d169b240e1aa2a
Author(s)
Lin, Ji
Liu, Zhijian
Wang, Hanrui
Han, Song
Date Issued
October 2018
Journal
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Publisher
Springer Science and Business Media LLC
Citation
He, Yihui et al. "AMC: AutoML for Model Compression and Acceleration on Mobile Devices." Computer vision -- ECCV 2018 : 15th European Conference, Lecture Notes in Computer Science, 11211, Springer, 2018, 815-832 © 2018 The Author(s)
Version
Author's final manuscript
Abstract
Model compression is an effective technique to efficiently deploy neural network models on mobile devices which have limited computation resources and tight power budgets. Conventional model compression techniques rely on hand-crafted features and require domain experts to explore the large design space trading off among model size, speed, and accuracy, which is usually sub-optimal and time-consuming. In this paper, we propose AutoML for Model Compression (AMC) which leverages reinforcement learning to efficiently sample the design space and can improve the model compression quality. We achieved state-of-the-art model compression results in a fully automated way without any human efforts. Under 4 × FLOPs reduction, we achieved 2.7% better accuracy than the hand-crafted model compression method for VGG-16 on ImageNet. We applied this automated, push-the-button compression pipeline to MobileNet-V1 and achieved a speedup of 1.53 × on the GPU (Titan Xp) and 1.95 × on an Android phone (Google Pixel 1), with negligible loss of accuracy.
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/10.1007/978-3-030-01234-2_48
