Off-policy reinforcement learning for efficient and effective GAN architecture search

• Main Authors: YUAN, Tian, QIN, Wang, HUANG, Zhiwu, LI, Wen, DAI, Dengxin, YANG, Minghao, WANG, Jun, FINK, Olga
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2020
• Subjects: Generative adversarial networks; Markov decision process; Neural architecture search; Off-policy; Reinforcement learning; Artificial Intelligence and Robotics; Systems Architecture
• Online Access: https://ink.library.smu.edu.sg/sis_research/6258; https://ink.library.smu.edu.sg/context/sis_research/article/7261/viewcontent/Off_PolicyReinforcementLearnin.pdf
• Institution: Singapore Management University

In this paper, we introduce a new reinforcement learning (RL) based neural architecture search (NAS) methodology for effective and efficient generative adversarial network (GAN) architecture search. The key idea is to formulate the GAN architecture search problem as a Markov decision process (MDP) for smoother architecture sampling, which enables a more effective RL-based search algorithm by targeting the potential global optimal architecture. To improve efficiency, we exploit an off-policy GAN architecture search algorithm that makes efficient use of the samples generated by previous policies. Evaluation on two standard benchmark datasets (i.e., CIFAR-10 and STL-10) demonstrates that the proposed method is able to discover highly competitive architectures for generally better image generation results with a considerably reduced computational burden: 7 GPU hours.