Artemis: HE-Aware Training for Efficient Privacy-Preserving Machine Learning. (arXiv:2310.01664v1 [cs.LG])

Privacy-Preserving ML (PPML) based on Homomorphic Encryption (HE) is a
promising foundational privacy technology. Making it more practical requires
lowering its computational cost, especially, in handling modern large deep
neural networks. Model compression via pruning is highly effective in
conventional plaintext ML but cannot be effectively applied to HE-PPML as is.

We propose Artemis, a highly effective DNN pruning technique for HE-based
inference. We judiciously investigate two HE-aware pruning strategies
(positional and diagonal) to reduce the number of Rotation operations, …

artemis aware compression computational cost effectively encryption handling homomorphic encryption large machine machine learning making networks neural networks plaintext privacy privacy technology technology training