Anti-Neuron Watermarking: Protecting Personal Data Against Unauthorized Neural Networks
This addresses data ownership protection for individual users against unauthorized neural network training, presenting a novel approach in this domain.
The paper tackles the problem of protecting individual users' image data from unauthorized use in training neural networks, especially when the data is a small fraction of the training set, by showing that watermarking images with a linear color transformation imprints a signature in the classifier, enabling third-party verification with specific properties.
We study protecting a user's data (images in this work) against a learner's unauthorized use in training neural networks. It is especially challenging when the user's data is only a tiny percentage of the learner's complete training set. We revisit the traditional watermarking under modern deep learning settings to tackle the challenge. We show that when a user watermarks images using a specialized linear color transformation, a neural network classifier will be imprinted with the signature so that a third-party arbitrator can verify the potentially unauthorized usage of the user data by inferring the watermark signature from the neural network. We also discuss what watermarking properties and signature spaces make the arbitrator's verification convincing. To our best knowledge, this work is the first to protect an individual user's data ownership from unauthorized use in training neural networks.