MONAS: Multi-Objective Neural Architecture Search using Reinforcement Learning

Recent studies on neural architecture search have shown that automatically designed neural networks perform as good as expert-crafted architectures. While most existing works aim at finding architectures that optimize the prediction accuracy, these architectures may have complexity and is therefore...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Chi-Hung, Hsu, Shu-Huan Chang, Jhao-Hong Liang, Chou, Hsin-Ping, Chun-Hao, Liu, Chang, Shih-Chieh, Jia-Yu, Pan, Yu-Ting, Chen, Wei, Wei, Da-Cheng, Juan
Format Paper
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 03.12.2018
Subjects
Accuracy
Computer architecture
Computer vision
Multiple objective analysis
Neural networks
Power consumption
Searching
Online AccessGet full text

Cover

More Information
Summary:Recent studies on neural architecture search have shown that automatically designed neural networks perform as good as expert-crafted architectures. While most existing works aim at finding architectures that optimize the prediction accuracy, these architectures may have complexity and is therefore not suitable being deployed on certain computing environment (e.g., with limited power budgets). We propose MONAS, a framework for Multi-Objective Neural Architectural Search that employs reward functions considering both prediction accuracy and other important objectives (e.g., power consumption) when searching for neural network architectures. Experimental results showed that, compared to the state-ofthe-arts, models found by MONAS achieve comparable or better classification accuracy on computer vision applications, while satisfying the additional objectives such as peak power.
ISSN:2331-8422