Absum: Simple regularization method for reducing structural sensitivity of convolutional neural networks

Sekitoshi Kanai, Yasutoshi Ida, Yasuhiro Fujiwara, Masanori Yamada, Shuichi Adachi

研究成果: Conference contribution

抄録

We propose Absum, which is a regularization method for improving adversarial robustness of convolutional neural networks (CNNs). Although CNNs can accurately recognize images, recent studies have shown that the convolution operations in CNNs commonly have structural sensitivity to specific noise composed of Fourier basis functions. By exploiting this sensitivity, they proposed a simple black-box adversarial attack: Single Fourier attack. To reduce structural sensitivity, we can use regularization of convolution filter weights since the sensitivity of linear transform can be assessed by the norm of the weights. However, standard regularization methods can prevent minimization of the loss function because they impose a tight constraint for obtaining high robustness. To solve this problem, Absum imposes a loose constraint; it penalizes the absolute values of the summation of the parameters in the convolution layers. Absum can improve robustness against single Fourier attack while being as simple and efficient as standard regularization methods (e.g., weight decay and L1 regularization). Our experiments demonstrate that Absum improves robustness against single Fourier attack more than standard regularization methods. Furthermore, we reveal that robust CNNs with Absum are more robust against transferred attacks due to decreasing the common sensitivity and against high-frequency noise than standard regularization methods. We also reveal that Absum can improve robustness against gradient-based attacks (projected gradient descent) when used with adversarial training.

本文言語English
ホスト出版物のタイトルAAAI 2020 - 34th AAAI Conference on Artificial Intelligence
出版社AAAI press
ページ4394-4403
ページ数10
ISBN(電子版)9781577358350
出版ステータスPublished - 2020
イベント34th AAAI Conference on Artificial Intelligence, AAAI 2020 - New York, United States
継続期間: 2020 2月 72020 2月 12

出版物シリーズ

名前AAAI 2020 - 34th AAAI Conference on Artificial Intelligence

Conference

Conference34th AAAI Conference on Artificial Intelligence, AAAI 2020
国/地域United States
CityNew York
Period20/2/720/2/12

ASJC Scopus subject areas

  • 人工知能

フィンガープリント

「Absum: Simple regularization method for reducing structural sensitivity of convolutional neural networks」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル