Back to Search Start Over

Similarity of Neural Architectures using Adversarial Attack Transferability

Authors :
Hwang, Jaehui
Han, Dongyoon
Heo, Byeongho
Park, Song
Chun, Sanghyuk
Lee, Jong-Seok
Publication Year :
2022

Abstract

In recent years, many deep neural architectures have been developed for image classification. Whether they are similar or dissimilar and what factors contribute to their (dis)similarities remains curious. To address this question, we aim to design a quantitative and scalable similarity measure between neural architectures. We propose Similarity by Attack Transferability (SAT) from the observation that adversarial attack transferability contains information related to input gradients and decision boundaries widely used to understand model behaviors. We conduct a large-scale analysis on 69 state-of-the-art ImageNet classifiers using our proposed similarity function to answer the question. Moreover, we observe neural architecture-related phenomena using model similarity that model diversity can lead to better performance on model ensembles and knowledge distillation under specific conditions. Our results provide insights into why developing diverse neural architectures with distinct components is necessary.<br />Comment: ECCV 2024; 35pages, 2.56MB

Details

Database :
arXiv
Publication Type :
Report
Accession number :
edsarx.2210.11407
Document Type :
Working Paper