Training DNN Model with Secret Key for Model Protection

TL;DR

This paper introduces a novel model protection technique using secret key-based pixel shuffling of input images, ensuring model security against various attacks while maintaining high accuracy with the correct key.

Contribution

The paper presents a new method for DNN model protection that employs block-wise pixel shuffling with a secret key during training, enhancing security without sacrificing performance.

Findings

Protected model accuracy is close to non-protected models with correct key.

Model security is robust against brute-force and fine-tuning attacks.

Accuracy drops significantly with incorrect key.

Abstract

In this paper, we propose a model protection method by using block-wise pixel shuffling with a secret key as a preprocessing technique to input images for the first time. The protected model is built by training with such preprocessed images. Experiment results show that the performance of the protected model is close to that of non-protected models when the key is correct, while the accuracy is severely dropped when an incorrect key is given, and the proposed model protection is robust against not only brute-force attacks but also fine-tuning attacks, while maintaining almost the same performance accuracy as that of using a non-protected model.