English

On the Effectiveness of Defensive Distillation

Cryptography and Security 2016-07-19 v1
Authors: Nicolas Papernot , Patrick McDaniel
View on arXiv ↗ PDF ↗

Abstract

We report experimental results indicating that defensive distillation successfully mitigates adversarial samples crafted using the fast gradient sign method, in addition to those crafted using the Jacobian-based iterative attack on which the defense mechanism was originally evaluated.

Keywords

defensive distillation adversarial examples fast gradient sign method jacobian-based iterative attack model robustness machine learning security

Cite

@article{arxiv.1607.05113,
  title  = {On the Effectiveness of Defensive Distillation},
  author = {Nicolas Papernot and Patrick McDaniel},
  journal= {arXiv preprint arXiv:1607.05113},
  year   = {2016}
}

Comments

Technical Report

Related papers

View all related →
Cryptography and Security · Computer Science
Distillation as a Defense to Adversarial Perturbations against Deep Neural Networks
Nicolas Papernot, Patrick McDaniel, Xi Wu, Somesh Jha +1
2016-03-15
Machine Learning · Computer Science
Extending Defensive Distillation
Nicolas Papernot, Patrick McDaniel
2017-05-16
Machine Learning · Computer Science
Denoising Autoencoder-based Defensive Distillation as an Adversarial Robustness Algorithm
Bakary Badjie, José Cecílio, António Casimiro
2023-03-29
Computer Vision and Pattern Recognition · Computer Science
Improving Defensive Distillation using Teacher Assistant
Maniratnam Mandal, Suna Gao
2023-05-16
Computation and Language · Computer Science
Evaluating Defensive Distillation For Defending Text Processing Neural Networks Against Adversarial Examples
Marcus Soll, Tobias Hinz, Sven Magg, Stefan Wermter
2019-08-22
Computer Vision and Pattern Recognition · Computer Science
Feature Distillation: DNN-Oriented JPEG Compression Against Adversarial Examples
Zihao Liu, Qi Liu, Tao Liu, Nuo Xu +3
2019-04-17
Machine Learning · Computer Science
Why Blocking Targeted Adversarial Perturbations Impairs the Ability to Learn
Ziv Katzir, Yuval Elovici
2019-07-15
Cryptography and Security · Computer Science
Evaluating Adversarial Robustness: A Comparison Of FGSM, Carlini-Wagner Attacks, And The Role of Distillation as Defense Mechanism
Trilokesh Ranjan Sarkar, Nilanjan Das, Pralay Sankar Maitra, Bijoy Some +4
2024-04-08
Cryptography and Security · Computer Science
Defensive Distillation is Not Robust to Adversarial Examples
Nicholas Carlini, David Wagner
2016-07-18
Cryptography and Security · Computer Science
Towards Evaluating the Robustness of Neural Networks
Nicholas Carlini, David Wagner
2017-03-23
Machine Learning · Computer Science
The Distillation Game: Adaptive Attacks & Efficient Defenses
Youssef Allouah, Mahdi Haghifam, Sanmi Koyejo, Reza Shokri
2026-05-29
Cryptography and Security · Computer Science
Adequacy of the Gradient-Descent Method for Classifier Evasion Attacks
Yi Han, Benjamin I. P. Rubinstein
2017-05-26
Computer Vision and Pattern Recognition · Computer Science
Distillation-Enhanced Physical Adversarial Attacks
Wei Liu, Yonglin Wu, Chaoqun Li, Zhuodong Liu +1
2025-01-07
Signal Processing · Electrical Eng. & Systems
Defending Against Adversarial Attack in ECG Classification with Adversarial Distillation Training
Jiahao Shao, Shijia Geng, Zhaoji Fu, Weilun Xu +2
2022-03-18
Cryptography and Security · Computer Science
BeDKD: Backdoor Defense Based on Directional Mapping Module and Adversarial Knowledge Distillation
Zhengxian Wu, Juan Wen, Wanli Peng, Yinghan Zhou +2
2026-01-15
Computer Vision and Pattern Recognition · Computer Science
A Statistical Defense Approach for Detecting Adversarial Examples
Alessandro Cennamo, Ido Freeman, Anton Kummert
2019-08-27
Machine Learning · Computer Science
Improving the Transferability of Adversarial Examples by Inverse Knowledge Distillation
Wenyuan Wu, Zheng Liu, Yong Chen, Chao Su +2
2025-02-25
Machine Learning · Computer Science
Rethinking Backdoor Attacks on Dataset Distillation: A Kernel Method Perspective
Ming-Yu Chung, Sheng-Yen Chou, Chia-Mu Yu, Pin-Yu Chen +2
2025-06-03
Machine Learning · Computer Science
Can Distillation Mitigate Backdoor Attacks in Pre-trained Encoders?
TIngxu Han, Wei Song, Weisong Sun, Ziqi Ding +6
2026-02-02
Machine Learning · Computer Science
Efficient Defenses Against Adversarial Attacks
Valentina Zantedeschi, Maria-Irina Nicolae, Ambrish Rawat
2017-08-31
Machine Learning · Computer Science
DROP: Poison Dilution via Knowledge Distillation for Federated Learning
Georgios Syros, Anshuman Suri, Farinaz Koushanfar, Cristina Nita-Rotaru +1
2025-04-29
Computer Vision and Pattern Recognition · Computer Science
Enhanced Attacks on Defensively Distilled Deep Neural Networks
Yujia Liu, Weiming Zhang, Shaohua Li, Nenghai Yu
2017-11-17
Cryptography and Security · Computer Science
INK: Inheritable Natural Backdoor Attack Against Model Distillation
Xiaolei Liu, Ming Yi, Kangyi Ding, Bangzhou Xin +3
2024-09-10
Artificial Intelligence · Computer Science
Antidistillation Sampling
Yash Savani, Asher Trockman, Zhili Feng, Yixuan Even Xu +4
2025-10-28
Machine Learning · Computer Science
A Computationally Efficient Method for Defending Adversarial Deep Learning Attacks
Rajeev Sahay, Rehana Mahfuz, Aly El Gamal
2019-06-14