xMLP: Revolutionizing Private Inference with Exclusive Square Activation

arXiv:2403.08024v1 Announce Type: cross
Abstract: Private Inference (PI) enables deep neural networks (DNNs) to work on private data without leaking sensitive information by exploiting cryptographic primitives such as multi-party computation (MPC) and homomorphic encryption (HE). However, the use of non-linear activations such as ReLU in DNNs can lead to impractically high PI latency in existing PI systems, as ReLU requires the use of costly MPC computations, such as Garbled Circuits. Since square activations can be processed by Beaver's triples hundreds …

arxiv can computation cryptographic cs.cr cs.lg data encryption exclusive exploiting high homomorphic encryption information linear mpc networks neural networks non party private private data sensitive sensitive information square work