Neural Network Training With Homomorphic Encryption
This work addresses the critical problem of privacy-preserving machine learning for users and model owners concerned about data and model confidentiality, representing an incremental step in applying homomorphic encryption to neural network training.
This paper introduces a method to train neural networks while preserving the confidentiality of both the model and the data, utilizing homomorphic encryption. The procedure is optimized for operations on packed ciphertexts, achieving a significant reduction in computations for multiplications and rotations from feedforward to back-propagation networks. The method was tested on the Iris dataset using the CKKS scheme with Microsoft SEAL.
We introduce a novel method and implementation architecture to train neural networks which preserves the confidentiality of both the model and the data. Our method relies on homomorphic capability of lattice based encryption scheme. Our procedure is optimized for operations on packed ciphertexts in order to achieve efficient updates of the model parameters. Our method achieves a significant reduction of computations due to our way to perform multiplications and rotations on packed ciphertexts from a feedforward network to a back-propagation network. To verify the accuracy of the training model as well as the implementation feasibility, we tested our method on the Iris data set by using the CKKS scheme with Microsoft SEAL as a back end. Although our test implementation is for simple neural network training, we believe our basic implementation block can help the further applications for more complex neural network based use cases.