Protecting Proprietary Data: Poisoning for Secure Dataset Release

Liam H Fowl, Ping-yeh Chiang, Micah Goldblum, Jonas Geiping, Arpit Amit Bansal, Wojciech Czaja, Tom Goldstein

Keywords: Secure Dataset Release, Data Poisoning, Availability Attack

Abstract: Large organizations such as social media companies continually release data, for example user images. At the same time, these organizations leverage their massive corpora of released data to train proprietary models that give them an edge over their competitors. These two behaviors can be in conflict as an organization wants to prevent competitors from using their own data to replicate the performance of their proprietary models. We solve this problem by developing a data poisoning method by which publicly released data can be minimally modified to prevent others from training models on it. Moreover, our method can be used in an online fashion so that companies can protect their data in real time as they release it. We demonstrate the success of our approach on ImageNet classification and on facial recognition.