Improving Neural Architecture Search by Minimizing Worst-Case Validation Loss
Zhang, Haochen Du, Xuefeng Zhang, Ruisi Xie, Pengtao
Zhang, Haochen
Du, Xuefeng
Zhang, Ruisi
Xie, Pengtao
Supervisor
Department
Machine Learning
Embargo End Date
Type
Journal article
Date
License
Language
English
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
Neural architecture search (NAS) aims at automatically searching for high-performance architectures and has achieved considerable progress. Existing NAS methods learn architectures by minimizing average-case validation losses. As a result, the searched architectures are less capable of making correct predictions under worst-case scenarios. To address this problem, we propose a framework which leverages a deep generative model to generate adversarial validation examples to measure the worst-case validation performance of an architecture and improves the architecture by minimizing the loss on the generated adversarial validation data. Our framework is based on multi-level optimization, which performs multiple learning stages end-to-end. Experiments on a variety of datasets demonstrate the effectiveness of our method.
Citation
H. Zhang, X. Du, R. Zhang, P. Xie, "Improving Neural Architecture Search by Minimizing Worst-Case Validation Loss," IEEE Transactions on Artificial Intelligence, vol. PP, no. 99, pp. 1-14, 2026, https://doi.org/10.1109/tai.2026.3670777.
Source
IEEE Transactions on Artificial Intelligence
Conference
Keywords
46 Information and Computing Sciences, 4611 Machine Learning
Subjects
Source
Publisher
IEEE