Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

REINFORCING THE ROBUSTNESS OF A DEEP NEURAL NETWORK TO ADVERSARIAL EXAMPLES BY USING COLOR QUANTIZATION OF TRAINING IMAGE DATA

Abstract: 

Recent works have shown the vulnerability of deep convolu-tional neural network (DCNN) to adversarial examples withmalicious perturbations. In particular, Black-Box attackswithout information of parameter and architectures of thetarget models are feared as realistic threats. To address thisproblem, we propose a method using an ensemble of mod-els trained by color-quantized data with loss maximization.Color-quantization can allow the trained models to focuson learning conspicuous spatial features to enhance the ro-bustness of DCNNs to adversarial examples. The proposedmethod can be adapted to Black-Box attacks with no needof particular attack algorithm for the defense. The resultsof our experiments validated the effectiveness for preventingdecrease in the test accuracy with adversarial perturbation.

up
0 users have voted:

Paper Details

Authors:
Xueting Wang, Toshihiko Yamasaki and Kiyoharu Aizawa
Submitted On:
20 September 2019 - 11:33am
Short Link:
Type:
Poster
Event:
Presenter's Name:
Shuntaro Miyazato
Paper Code:
3045
Document Year:
2019
Cite

Document Files

ICIP.pdf

(9)

Subscribe

[1] Xueting Wang, Toshihiko Yamasaki and Kiyoharu Aizawa, "REINFORCING THE ROBUSTNESS OF A DEEP NEURAL NETWORK TO ADVERSARIAL EXAMPLES BY USING COLOR QUANTIZATION OF TRAINING IMAGE DATA", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4769. Accessed: Oct. 18, 2019.
@article{4769-19,
url = {http://sigport.org/4769},
author = {Xueting Wang; Toshihiko Yamasaki and Kiyoharu Aizawa },
publisher = {IEEE SigPort},
title = {REINFORCING THE ROBUSTNESS OF A DEEP NEURAL NETWORK TO ADVERSARIAL EXAMPLES BY USING COLOR QUANTIZATION OF TRAINING IMAGE DATA},
year = {2019} }
TY - EJOUR
T1 - REINFORCING THE ROBUSTNESS OF A DEEP NEURAL NETWORK TO ADVERSARIAL EXAMPLES BY USING COLOR QUANTIZATION OF TRAINING IMAGE DATA
AU - Xueting Wang; Toshihiko Yamasaki and Kiyoharu Aizawa
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4769
ER -
Xueting Wang, Toshihiko Yamasaki and Kiyoharu Aizawa. (2019). REINFORCING THE ROBUSTNESS OF A DEEP NEURAL NETWORK TO ADVERSARIAL EXAMPLES BY USING COLOR QUANTIZATION OF TRAINING IMAGE DATA. IEEE SigPort. http://sigport.org/4769
Xueting Wang, Toshihiko Yamasaki and Kiyoharu Aizawa, 2019. REINFORCING THE ROBUSTNESS OF A DEEP NEURAL NETWORK TO ADVERSARIAL EXAMPLES BY USING COLOR QUANTIZATION OF TRAINING IMAGE DATA. Available at: http://sigport.org/4769.
Xueting Wang, Toshihiko Yamasaki and Kiyoharu Aizawa. (2019). "REINFORCING THE ROBUSTNESS OF A DEEP NEURAL NETWORK TO ADVERSARIAL EXAMPLES BY USING COLOR QUANTIZATION OF TRAINING IMAGE DATA." Web.
1. Xueting Wang, Toshihiko Yamasaki and Kiyoharu Aizawa. REINFORCING THE ROBUSTNESS OF A DEEP NEURAL NETWORK TO ADVERSARIAL EXAMPLES BY USING COLOR QUANTIZATION OF TRAINING IMAGE DATA [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4769