preprints.org > computer science and mathematics > artificial intelligence and machine learning > doi: 10.20944/preprints202303.0345.v1

Preprint Article Version 1 Preserved in Portico This version is not peer-reviewed

Version 1 : Received: 17 March 2023 / Approved: 20 March 2023 / Online: 20 March 2023 (06:08:35 CET)
Version 2 : Received: 28 March 2023 / Approved: 29 March 2023 / Online: 29 March 2023 (02:14:14 CEST)

Existing one-shot neural architecture search (NAS) methods have to conduct a search over a giant super-net, which leads to the huge computational cost. To reduce such cost, in this paper, we propose a method, called FTSO, to divide the whole architecture search into two sub-steps. Specifically, in the first step, we only search for the topology, and in the second step, we search for the operators. FTSO not only reduces NAS’s search time from days to 0.68 seconds, but also significantly improves the found architecture's accuracy. Our extensive experiments on ImageNet show that within 18 seconds, FTSO can achieve a 76.4% testing accuracy, 1.5% higher than the SOTA, PC-DARTS. In addition, FTSO can reach a 97.77% testing accuracy, 0.27% higher than the SOTA, with nearly 100% (99.8%) search time saved, when searching on CIFAR10.

neural architecture search; machine learning; computer vision

Computer Science and Mathematics, Artificial Intelligence and Machine Learning