PrivCirNet: Efficient Private Inference via Block Circulant Transformation

arXiv:2405.14569v1 Announce Type: new
Abstract: Homomorphic encryption (HE)-based deep neural network (DNN) inference protects data and model privacy but suffers from significant computation overhead. We observe transforming the DNN weights into circulant matrices converts general matrix-vector multiplications into HE-friendly 1-dimensional convolutions, drastically reducing the HE computation cost. Hence, in this paper, we propose \method, a protocol/network co-optimization framework based on block circulant transformation. At the protocol level, PrivCirNet customizes the HE encoding algorithm that is fully compatible with the block …

arxiv block computation cost cs.ai cs.cr data encryption general homomorphic encryption matrix network neural network observe privacy private transformation