Bi-fidelity evolutionary multiobjective search for adversarially robust deep neural architectures

Liu，Jia1; Cheng，Ran2; Jin，Yaochu1,3

2023-09-14

10.1016/j.neucom.2023.126465

Neurocomputing   Impact Factor & Quartile

0925-2312

1872-8286

Deep neural networks have been found vulnerable to adversarial attacks, thus raising potential concerns in security-sensitive contexts. To address this problem, recent research has investigated the adversarial robustness of deep neural networks from the architectural point of view. However, searching for architectures of deep neural networks is computationally expensive, particularly when coupled with an adversarial training process. To meet the above challenge, this paper proposes a bi-fidelity multiobjective neural architecture search approach. First, we formulate the neural architecture search (NAS) problem for enhancing the adversarial robustness of deep neural networks into a multiobjective optimization problem. Specifically, in addition to using low-fidelity estimations as the primary objectives, we leverage the output of a surrogate model trained with high-fidelity evaluations as an auxiliary objective. Secondly, we reduce the computational cost by combining three performance estimation methods, i.e., parameter sharing, low-fidelity evaluation, and surrogate-based predictor. The effectiveness of the proposed approach is confirmed by extensive experiments conducted on CIFAR-10, CIFAR-100 and SVHN datasets.

Adversarial attacks Low-fidelity Multiobjectivization Neural architecture search Surrogate

SCI

English

Others

Computer Science

Computer Science, Artificial Intelligence

WOS:001035085900001

ELSEVIER

COMPUTER SCIENCE

2-s2.0-85164294065

Scopus

1.Department of Computer Science,University of Surrey,Guildford,United Kingdom
2.Department of Computer Science and Engineering,Southern University of Science and Technology,Shenzhen,China
3.Faculty of Technology,Bielefeld University,Bielefeld,Germany

Liu，Jia,Cheng，Ran,Jin，Yaochu. Bi-fidelity evolutionary multiobjective search for adversarially robust deep neural architectures[J]. Neurocomputing,2023,550.

Liu，Jia,Cheng，Ran,&Jin，Yaochu.(2023).Bi-fidelity evolutionary multiobjective search for adversarially robust deep neural architectures.Neurocomputing,550.

Liu，Jia,et al."Bi-fidelity evolutionary multiobjective search for adversarially robust deep neural architectures".Neurocomputing 550(2023).