Robust Watermarking of Neural Network with Exponential Weighting

Abstract

Deep learning has been achieving top performance in many tasks. Since training of a deep learning model requires a great deal of cost, we need to treat neural network models as valuable intellectual properties. One concern in such a situation is that some malicious user might redistribute the model or provide a prediction service using the model without permission. One promising solution is digital watermarking, to embed a mechanism into the model so that the owner of the model can verify the ownership of the model externally. In this study, we present a novel attack method against watermark, query modification, and demonstrate that all of the existing watermark methods are vulnerable to either of query modification or existing attack method (model modification). To overcome this vulnerability, we present a novel watermarking method, exponential weighting. We experimentally show that our watermarking method achieves high verification performance of watermark even under a malicious attempt of unauthorized service providers, such as model modification and query modification, without sacrificing the predictive performance of the neural network model.

arXiv comment13 pages

References in corpus (7)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• How transferable are features in deep neural networks?
• Deep Speech 2: End-to-End Speech Recognition in English and Mandarin
• Embedding Watermarks into Deep Neural Networks
• EMNIST: an extension of MNIST to handwritten letters
• Adversarial Frontier Stitching for Remote Neural Network Watermarking
• DeepSigns: A Generic Watermarking Framework for IP Protection of Deep Learning Models

Cited by in corpus (5)

• How to Prove Your Model Belongs to You: A Blind-Watermark based Framework to Protect Intellectual Property of DNN
• Towards Practical Watermark for Deep Neural Networks in Federated Learning
• MimosaNet: An Unrobust Neural Network Preventing Model Stealing
• Proof-of-Learning: Definitions and Practice
• WAFFLE: Watermarking in Federated Learning