C2PI: An Efficient Crypto-Clear Two-Party Neural Network Private Inference. (arXiv:2304.13266v1 [cs.CR])

Recently, private inference (PI) has addressed the rising concern over data
and model privacy in machine learning inference as a service. However, existing
PI frameworks suffer from high computational and communication costs due to the
expensive multi-party computation (MPC) protocols. Existing literature has
developed lighter MPC protocols to yield more efficient PI schemes. We, in
contrast, propose to lighten them by introducing an empirically-defined privacy
evaluation. To that end, we reformulate the threat model of PI and use
inference data …

attacks communication computation computational crypto data data privacy defined end evaluation frameworks high literature machine machine learning mpc network neural network party privacy private protocols rising service threat threat model