On the Relationship between Generalization and Robustness to Adversarial Examples

被引:8
|
作者
Pedraza, Anibal [1 ]
Deniz, Oscar [1 ]
Bueno, Gloria [1 ]
机构
[1] Univ Castilla La Mancha, VISILAB, ETSII, Ciudad Real 13071, Spain
来源
SYMMETRY-BASEL | 2021年 / 13卷 / 05期
关键词
machine learning; computer vision; deep learning; adversarial examples; adversarial robustness; overfitting;
D O I
10.3390/sym13050817
中图分类号
O [数理科学和化学]; P [天文学、地球科学]; Q [生物科学]; N [自然科学总论];
学科分类号
07 ; 0710 ; 09 ;
摘要
One of the most intriguing phenomenons related to deep learning is the so-called adversarial examples. These samples are visually equivalent to normal inputs, undetectable for humans, yet they cause the networks to output wrong results. The phenomenon can be framed as a symmetry/asymmetry problem, whereby inputs to a neural network with a similar/symmetric appearance to regular images, produce an opposite/asymmetric output. Some researchers are focused on developing methods for generating adversarial examples, while others propose defense methods. In parallel, there is a growing interest in characterizing the phenomenon, which is also the focus of this paper. From some well known datasets of common images, like CIFAR-10 and STL-10, a neural network architecture is first trained in a normal regime, where training and validation performances increase, reaching generalization. Additionally, the same architectures and datasets are trained in an overfitting regime, where there is a growing disparity in training and validation performances. The behaviour of these two regimes against adversarial examples is then compared. From the results, we observe greater robustness to adversarial examples in the overfitting regime. We explain this simultaneous loss of generalization and gain in robustness to adversarial examples as another manifestation of the well-known fitting-generalization trade-off.
引用
收藏
页数:13
相关论文
共 50 条
  • [1] ROBUSTNESS OF DEEP NEURAL NETWORKS IN ADVERSARIAL EXAMPLES
    Teng, Da
    Song, Xiao m
    Gong, Guanghong
    Han, Liang
    INTERNATIONAL JOURNAL OF INDUSTRIAL ENGINEERING-THEORY APPLICATIONS AND PRACTICE, 2017, 24 (02): : 123 - 133
  • [2] Robustness to adversarial examples can be improved with overfitting
    Deniz, Oscar
    Pedraza, Anibal
    Vallez, Noelia
    Salido, Jesus
    Bueno, Gloria
    INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS, 2020, 11 (04) : 935 - 944
  • [3] Regularizing Hard Examples Improves Adversarial Robustness
    Lee, Hyungyu
    Lee, Saehyung
    Bae, Ho
    Yoon, Sungroh
    JOURNAL OF MACHINE LEARNING RESEARCH, 2025, 26
  • [4] Robustness to adversarial examples can be improved with overfitting
    Oscar Deniz
    Anibal Pedraza
    Noelia Vallez
    Jesus Salido
    Gloria Bueno
    International Journal of Machine Learning and Cybernetics, 2020, 11 : 935 - 944
  • [5] Toward Enhanced Adversarial Robustness Generalization in Object Detection: Feature Disentangled Domain Adaptation for Adversarial Training
    Jung, Yoojin
    Song, Byung Cheol
    IEEE ACCESS, 2024, 12 : 179065 - 179076
  • [6] Analyzing the Robustness of Deep Learning Against Adversarial Examples
    Zhao, Jun
    2018 56TH ANNUAL ALLERTON CONFERENCE ON COMMUNICATION, CONTROL, AND COMPUTING (ALLERTON), 2018, : 1060 - 1064
  • [7] Adversarial Minimax Training for Robustness Against Adversarial Examples
    Komiyama, Ryota
    Hattori, Motonobu
    NEURAL INFORMATION PROCESSING (ICONIP 2018), PT II, 2018, 11302 : 690 - 699
  • [8] Effect of adversarial examples on the robustness of CAPTCHA
    Zhang, Yang
    Gao, Haichang
    Pei, Ge
    Kang, Shuai
    Zhou, Xin
    2018 INTERNATIONAL CONFERENCE ON CYBER-ENABLED DISTRIBUTED COMPUTING AND KNOWLEDGE DISCOVERY (CYBERC 2018), 2018, : 1 - 10
  • [9] Attack as Defense: Characterizing Adversarial Examples using Robustness
    Zhao, Zhe
    Chen, Guangke
    Wang, Jingyi
    Yang, Yiwei
    Song, Fu
    Sun, Jun
    ISSTA '21: PROCEEDINGS OF THE 30TH ACM SIGSOFT INTERNATIONAL SYMPOSIUM ON SOFTWARE TESTING AND ANALYSIS, 2021, : 42 - 55
  • [10] On the robustness of randomized classifiers to adversarial examples
    Pinot, Rafael
    Meunier, Laurent
    Yger, Florian
    Gouy-Pailler, Cedric
    Chevaleyre, Yann
    Atif, Jamal
    MACHINE LEARNING, 2022, 111 (09) : 3425 - 3457