We explore efficient neural architecture search methods and show that a
simple yet powerful evolutionary algorithm can discover new architectures with
excellent performance. Our approach combines a novel hierarchical genetic
representation scheme that imitates the modularized design pattern commonly
adopted by human experts, and an expressive search space that supports complex
topologies. Our algorithm efficiently discovers architectures that outperform a
large number of manually designed models for image classification, obtaining
top-1 error of 3.6% on CIFAR-10 and 20.3% when transferred to ImageNet, which
is competitive with the best existing neural architecture search approaches. We
also present results using random search, achieving 0.3% less top-1 accuracy on
CIFAR-10 and 0.1% less on ImageNet whilst reducing the search time from 36
hours down to 1 hour.
Description
Hierarchical Representations for Efficient Architecture Search
%0 Generic
%1 liu2017hierarchical
%A Liu, Hanxiao
%A Simonyan, Karen
%A Vinyals, Oriol
%A Fernando, Chrisantha
%A Kavukcuoglu, Koray
%D 2017
%K l2l to_read
%T Hierarchical Representations for Efficient Architecture Search
%U http://arxiv.org/abs/1711.00436
%X We explore efficient neural architecture search methods and show that a
simple yet powerful evolutionary algorithm can discover new architectures with
excellent performance. Our approach combines a novel hierarchical genetic
representation scheme that imitates the modularized design pattern commonly
adopted by human experts, and an expressive search space that supports complex
topologies. Our algorithm efficiently discovers architectures that outperform a
large number of manually designed models for image classification, obtaining
top-1 error of 3.6% on CIFAR-10 and 20.3% when transferred to ImageNet, which
is competitive with the best existing neural architecture search approaches. We
also present results using random search, achieving 0.3% less top-1 accuracy on
CIFAR-10 and 0.1% less on ImageNet whilst reducing the search time from 36
hours down to 1 hour.
@misc{liu2017hierarchical,
abstract = {We explore efficient neural architecture search methods and show that a
simple yet powerful evolutionary algorithm can discover new architectures with
excellent performance. Our approach combines a novel hierarchical genetic
representation scheme that imitates the modularized design pattern commonly
adopted by human experts, and an expressive search space that supports complex
topologies. Our algorithm efficiently discovers architectures that outperform a
large number of manually designed models for image classification, obtaining
top-1 error of 3.6% on CIFAR-10 and 20.3% when transferred to ImageNet, which
is competitive with the best existing neural architecture search approaches. We
also present results using random search, achieving 0.3% less top-1 accuracy on
CIFAR-10 and 0.1% less on ImageNet whilst reducing the search time from 36
hours down to 1 hour.},
added-at = {2018-02-26T11:14:07.000+0100},
author = {Liu, Hanxiao and Simonyan, Karen and Vinyals, Oriol and Fernando, Chrisantha and Kavukcuoglu, Koray},
biburl = {https://www.bibsonomy.org/bibtex/2943bfccf42cec2035ed12b3f353fc480/jk_itwm},
description = {Hierarchical Representations for Efficient Architecture Search},
interhash = {678c33107f5301735e18b17cbecf0c5a},
intrahash = {943bfccf42cec2035ed12b3f353fc480},
keywords = {l2l to_read},
note = {cite arxiv:1711.00436Comment: Accepted as a conference paper at ICLR 2018},
timestamp = {2018-02-26T11:14:07.000+0100},
title = {Hierarchical Representations for Efficient Architecture Search},
url = {http://arxiv.org/abs/1711.00436},
year = 2017
}