TY - GEN
T1 - Evolutionary Multi-objective Architecture Search Framework: Application to COVID-19 3D CT Classification
AU - He, Xin
AU - Ying, Guohao
AU - Zhang, Jiyong
AU - Chu, Xiaowen
N1 - RGC RMGS2019 1 23 (from Hong Kong Research Matching Grant Scheme)
Funding Information:
Acknowledgement. This work was supported in part by Hong Kong Research Matching Grant RMGS2019_1_23, the Zhejiang Province Nature Science Foundation of China under Grant LZ22F020003, and the HDU-CECDATA Joint Research Center of Big Data Technologies under Grant KYH063120009.
Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2022/9/15
Y1 - 2022/9/15
N2 - The COVID-19 pandemic has threatened global health. Many studies have applied deep convolutional neural networks (CNN) to recognize COVID-19 based on chest 3D computed tomography (CT). Recent works show that no model generalizes well across CT datasets from different countries, and manually designing models for specific datasets requires expertise; thus, neural architecture search (NAS) that aims to search models automatically has become an attractive solution. To reduce the search cost on large 3D CT datasets, most NAS-based works use the weight-sharing (WS) strategy to make all models share weights within a supernet; however, WS inevitably incurs search instability, leading to inaccurate model estimation. In this work, we propose an efficient Evolutionary Multi-objective ARchitecture Search (EMARS) framework. We propose a new objective, namely potential, which can help exploit promising models to indirectly reduce the number of models involved in weights training, thus alleviating search instability. We demonstrate that under objectives of accuracy and potential, EMARS can balance exploitation and exploration, i.e., reducing search time and finding better models. Our searched models are small and perform better than prior works on three public COVID-19 3D CT datasets.
AB - The COVID-19 pandemic has threatened global health. Many studies have applied deep convolutional neural networks (CNN) to recognize COVID-19 based on chest 3D computed tomography (CT). Recent works show that no model generalizes well across CT datasets from different countries, and manually designing models for specific datasets requires expertise; thus, neural architecture search (NAS) that aims to search models automatically has become an attractive solution. To reduce the search cost on large 3D CT datasets, most NAS-based works use the weight-sharing (WS) strategy to make all models share weights within a supernet; however, WS inevitably incurs search instability, leading to inaccurate model estimation. In this work, we propose an efficient Evolutionary Multi-objective ARchitecture Search (EMARS) framework. We propose a new objective, namely potential, which can help exploit promising models to indirectly reduce the number of models involved in weights training, thus alleviating search instability. We demonstrate that under objectives of accuracy and potential, EMARS can balance exploitation and exploration, i.e., reducing search time and finding better models. Our searched models are small and perform better than prior works on three public COVID-19 3D CT datasets.
KW - 3D Computed Tomograph (CT)
KW - COVID-19
KW - Evolutionary Algorithm (EA)
KW - Neural Architecture Search (NAS)
KW - Weight-sharing
UR - http://www.scopus.com/inward/record.url?scp=85138805221&partnerID=8YFLogxK
U2 - 10.48550/arXiv.2101.10667
DO - 10.48550/arXiv.2101.10667
M3 - Conference proceeding
SN - 9783031164309
T3 - Lecture Notes in Computer Science
SP - 560
EP - 570
BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2022
A2 - Wang, Linwei
A2 - Dou, Qi
A2 - Fletcher, P. Thomas
A2 - Speidel, Stefanie
A2 - Li, Shuo
PB - Springer Cham
T2 - 25th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2022
Y2 - 18 September 2022 through 22 September 2022
ER -