初始化
计算机科学
人工神经网络
渡线
人口
公制(单位)
人工智能
遗传算法
机器学习
网络体系结构
集合(抽象数据类型)
网(多面体)
进化算法
工程类
数学
计算机安全
运营管理
社会学
人口学
程序设计语言
几何学
作者
Zhichao Lu,Ian Whalen,Vishnu Naresh Boddeti,Yashesh Dhebar,Kalyanmoy Deb,Erik D. Goodman,Wolfgang Banzhaf
出处
期刊:Cornell University - arXiv
日期:2018-10-08
被引量:10
摘要
This paper introduces NSGA-Net -- an evolutionary approach for neural architecture search (NAS). NSGA-Net is designed with three goals in mind: (1) a procedure considering multiple and conflicting objectives, (2) an efficient procedure balancing exploration and exploitation of the space of potential neural network architectures, and (3) a procedure finding a diverse set of trade-off network architectures achieved in a single run. NSGA-Net is a population-based search algorithm that explores a space of potential neural network architectures in three steps, namely, a population initialization step that is based on prior-knowledge from hand-crafted architectures, an exploration step comprising crossover and mutation of architectures, and finally an exploitation step that utilizes the hidden useful knowledge stored in the entire history of evaluated neural architectures in the form of a Bayesian Network. Experimental results suggest that combining the dual objectives of minimizing an error metric and computational complexity, as measured by FLOPs, allows NSGA-Net to find competitive neural architectures. Moreover, NSGA-Net achieves error rate on the CIFAR-10 dataset on par with other state-of-the-art NAS methods while using orders of magnitude less computational resources. These results are encouraging and shows the promise to further use of EC methods in various deep-learning paradigms.
科研通智能强力驱动
Strongly Powered by AbleSci AI