Guardado en:
Detalles Bibliográficos
Autores principales: He, Yihui, Lin, Ji, Liu, Zhijian, Wang, Hanrui, Li, Li-Jia, Han, Song
Formato: Preprint
Publicado: 2018
Materias:
Acceso en línea:https://arxiv.org/abs/1802.03494
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917629684350976
author He, Yihui
Lin, Ji
Liu, Zhijian
Wang, Hanrui
Li, Li-Jia
Han, Song
author_facet He, Yihui
Lin, Ji
Liu, Zhijian
Wang, Hanrui
Li, Li-Jia
Han, Song
contents Model compression is a critical 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 heuristics and rule-based policies that 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 leverage reinforcement learning to provide the model compression policy. This learning-based compression policy outperforms conventional rule-based compression policy by having higher compression ratio, better preserving the accuracy and freeing human labor. Under 4x FLOPs reduction, we achieved 2.7% better accuracy than the handcrafted model compression policy for VGG-16 on ImageNet. We applied this automated, push-the-button compression pipeline to MobileNet and achieved 1.81x speedup of measured inference latency on an Android phone and 1.43x speedup on the Titan XP GPU, with only 0.1% loss of ImageNet Top-1 accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_1802_03494
institution arXiv
publishDate 2018
record_format arxiv
spellingShingle AMC: AutoML for Model Compression and Acceleration on Mobile Devices
He, Yihui
Lin, Ji
Liu, Zhijian
Wang, Hanrui
Li, Li-Jia
Han, Song
Computer Vision and Pattern Recognition
Model compression is a critical 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 heuristics and rule-based policies that 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 leverage reinforcement learning to provide the model compression policy. This learning-based compression policy outperforms conventional rule-based compression policy by having higher compression ratio, better preserving the accuracy and freeing human labor. Under 4x FLOPs reduction, we achieved 2.7% better accuracy than the handcrafted model compression policy for VGG-16 on ImageNet. We applied this automated, push-the-button compression pipeline to MobileNet and achieved 1.81x speedup of measured inference latency on an Android phone and 1.43x speedup on the Titan XP GPU, with only 0.1% loss of ImageNet Top-1 accuracy.
title AMC: AutoML for Model Compression and Acceleration on Mobile Devices
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/1802.03494