ICML 2022spotlight65 citations
Certified Neural Network Watermarks with Randomized Smoothing
Arpit Bansal, Ping-Yeh Chiang, Michael J Curry, Rajiv Jain, Curtis Wigington, Varun Manjunatha, John P Dickerson, Tom Goldstein
Abstract
Watermarking is a commonly used strategy to protect creators’ rights to digital images, videos and audio. Recently, watermarking methods have been extended to deep learning models – in principle, the watermark should be preserved when an adversary tries to copy the model. However, in practice, watermarks can often be removed by an intelligent adversary. Several papers have proposed watermarking methods that claim to be empirically resistant to different types of removal attacks, but these new techniques often fail in the face of new or better-tuned adversaries. In this paper, we propose the first
BibTeX
@InProceedings{pmlr-v162-bansal22a,
title = {Certified Neural Network Watermarks with Randomized Smoothing},
author = {Bansal, Arpit and Chiang, Ping-Yeh and Curry, Michael J and Jain, Rajiv and Wigington, Curtis and Manjunatha, Varun and Dickerson, John P and Goldstein, Tom},
booktitle = {Proceedings of the 39th International Conference on Machine Learning},
pages = {1450--1465},
year = {2022},
editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan},
volume = {162},
series = {Proceedings of Machine Learning Research},
month = {17--23 Jul},
publisher = {PMLR},
pdf = {https://proceedings.mlr.press/v162/bansal22a/bansal22a.pdf},
url = {https://proceedings.mlr.press/v162/bansal22a.html},
abstract = {Watermarking is a commonly used strategy to protect creators’ rights to digital images, videos and audio. Recently, watermarking methods have been extended to deep learning models – in principle, the watermark should be preserved when an adversary tries to copy the model. However, in practice, watermarks can often be removed by an intelligent adversary. Several papers have proposed watermarking methods that claim to be empirically resistant to different types of removal attacks, but these new techniques often fail in the face of new or better-tuned adversaries. In this paper, we propose the first