1/1
2 files

Protecting IP of deep neural networks with watermarking: a new label helps

Download all (1.14 MB)
conference contribution
posted on 2020-01-01, 00:00 authored by Qi ZhongQi Zhong, Leo ZhangLeo Zhang, J Zhang, Longxiang GaoLongxiang Gao, Yong XiangYong Xiang
Deep neural network (DNN) models have shown great success in almost every artificial area. It is a non-trivial task to build a good DNN model. Nowadays, various MLaaS providers have launched their cloud services, which trains DNN models for users. Once they are released, driven by potential monetary profit, the models may be duplicated, resold, or redistributed by adversaries, including greedy service providers themselves. To mitigate this threat, in this paper, we propose an innovative framework to protect the intellectual property of deep learning models, that is, watermarking the model by adding a new label to crafted key samples during training. The intuition comes from the fact that, compared with existing DNN watermarking methods, adding a new label will not twist the original decision boundary but can help the model learn the features of key samples better. We implement a prototype of our framework and evaluate the performance under three different benchmark datasets, and investigate the relationship between model accuracy, perturbation strength, and key samples’ length. Extensive experimental results show that, compared with the existing schemes, the proposed method performs better under small perturbation strength or short key samples’ length in terms of classification accuracy and ownership verification efficiency.

History

Event

Knowledge Discovery and Data Mining. Conference (24th : 2020 : Singapore)

Volume

12085

Series

Knowledge Discovery and Data Mining Conference

Pagination

462 - 474

Publisher

Springer

Location

Singapore

Place of publication

Cham, Switzerland

Start date

2020-05-11

End date

2020-05-14

ISSN

0302-9743

eISSN

1611-3349

ISBN-13

9783030474355

Language

eng

Publication classification

E1 Full written paper - refereed

Editor/Contributor(s)

H Lauw, R Wong, A Ntoulas, E Lim, S Ng, S Pan

Title of proceedings

PAKDD 2020 : Proceedings of the 24th Pacific-Asia Conference on Knowledge Discovery and Data Mining