Paper Info
Title
Sample-Efficient Neural Architecture Search by Learning Actions for Monte Carlo Tree Search
Abstract
Neural Architecture Search (NAS) has emerged as a promising technique for automatic neural network design. However, existing MCTS based NAS approaches often utilize manually designed action space, which is not directly related to the performance metric to be optimized (e.g., accuracy), leading to sample-inefficient explorations of architectures. To improve the sample efficiency, this paper proposes Latent Action Neural Architecture Search (LaNAS), which learns actions to recursively partition the search space into good or bad regions that contain networks with similar performance metrics. During the search phase, as different action sequences lead to regions with different performance, the search efficiency can be significantly improved by biasing towards the good regions. On three NAS tasks, empirical results demonstrate that LaNAS is at least an order more sample efficient than baseline methods including evolutionary algorithms, Bayesian optimizations, and random search. When applied in practice, both one-shot and regular LaNAS consistently outperform existing results. Particularly, LaNAS achieves 99.0 percent accuracy on CIFAR-10 and 80.8 percent top1 accuracy at 600 MFLOPS on ImageNet in only 800 samples, significantly outperforming AmoebaNet with <inline-formula><tex-math notation="LaTeX">$33\times$</tex-math></inline-formula> fewer samples. Our code is publicly available at <uri>https://github.com/facebookresearch/LaMCTS</uri> .
Year
DOI
Venue
2022
10.1109/TPAMI.2021.3071343
IEEE Transactions on Pattern Analysis and Machine Intelligence
Keywords
DocType
Volume
Algorithms,Bayes Theorem,Monte Carlo Method,Neural Networks, Computer
Journal
44
Issue
ISSN
Citations 
9
0162-8828
0
PageRank 
References 
Authors
0.34
7
5
Name
Order
Citations
PageRank
linnan wang1405.59
Saining Xie223112.45
Teng Li37221.44
Rodrigo Fonseca42390144.33
Yuandong Tian570343.06