An Encryption Method of ConvMixer Models without Performance Degradation
This work addresses model protection and privacy in image classification for AI security applications, representing an incremental improvement over existing encryption methods.
The paper tackles the problem of performance degradation in encrypted deep neural networks by proposing a novel encryption method for ConvMixer models that maintains classification accuracy when using encrypted test images with a secret key, achieving the same performance as plain models on the CIFAR10 dataset.
In this paper, we propose an encryption method for ConvMixer models with a secret key. Encryption methods for DNN models have been studied to achieve adversarial defense, model protection and privacy-preserving image classification. However, the use of conventional encryption methods degrades the performance of models compared with that of plain models. Accordingly, we propose a novel method for encrypting ConvMixer models. The method is carried out on the basis of an embedding architecture that ConvMixer has, and models encrypted with the method can have the same performance as models trained with plain images only when using test images encrypted with a secret key. In addition, the proposed method does not require any specially prepared data for model training or network modification. In an experiment, the effectiveness of the proposed method is evaluated in terms of classification accuracy and model protection in an image classification task on the CIFAR10 dataset.