TY - GEN
T1 - Simple Post-Training Robustness using Test Time Augmentations and Random Forest
AU - Cohen, Gilad
AU - Giryes, Raja
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024/1/3
Y1 - 2024/1/3
N2 - Although Deep Neural Networks (DNNs) achieve excellent performance on many real-world tasks, they are highly vulnerable to adversarial attacks. A leading defense against such attacks is adversarial training, a technique in which a DNN is trained to be robust to adversarial attacks by introducing adversarial noise to its input. This procedure is effective but must be done during the training phase. In this work, we propose Augmented Random Forest (ARF), a simple and easy-to-use strategy for robustifying an existing pretrained DNN without modifying its weights. For every image, we generate randomized test time augmentations by applying diverse color, blur, noise, and geometric transforms. Then we use the DNN's logits output to train a simple random forest to predict the real class label. Our method achieves state-of-the-art adversarial robustness on a diversity of white and black box attacks with minimal compromise on the natural images' classification. We test ARF also against numerous adaptive white-box attacks and it shows excellent results when combined with adversarial training. https://github.com/giladcohen/ARF.
AB - Although Deep Neural Networks (DNNs) achieve excellent performance on many real-world tasks, they are highly vulnerable to adversarial attacks. A leading defense against such attacks is adversarial training, a technique in which a DNN is trained to be robust to adversarial attacks by introducing adversarial noise to its input. This procedure is effective but must be done during the training phase. In this work, we propose Augmented Random Forest (ARF), a simple and easy-to-use strategy for robustifying an existing pretrained DNN without modifying its weights. For every image, we generate randomized test time augmentations by applying diverse color, blur, noise, and geometric transforms. Then we use the DNN's logits output to train a simple random forest to predict the real class label. Our method achieves state-of-the-art adversarial robustness on a diversity of white and black box attacks with minimal compromise on the natural images' classification. We test ARF also against numerous adaptive white-box attacks and it shows excellent results when combined with adversarial training. https://github.com/giladcohen/ARF.
KW - Adversarial learning
KW - Algorithms
KW - adversarial attack and defense methods
UR - http://www.scopus.com/inward/record.url?scp=85192021855&partnerID=8YFLogxK
U2 - 10.1109/WACV57701.2024.00395
DO - 10.1109/WACV57701.2024.00395
M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???
AN - SCOPUS:85192021855
T3 - Proceedings - 2024 IEEE Winter Conference on Applications of Computer Vision, WACV 2024
SP - 3984
EP - 3994
BT - Proceedings - 2024 IEEE Winter Conference on Applications of Computer Vision, WACV 2024
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2024 IEEE Winter Conference on Applications of Computer Vision, WACV 2024
Y2 - 4 January 2024 through 8 January 2024
ER -