Tabula: Efficiently Computing Nonlinear Activation Functions for Secure Neural Network Inference

arXiv:2203.02833v2 Announce Type: replace
Abstract: Multiparty computation approaches to secure neural network inference commonly rely on garbled circuits for securely executing nonlinear activation functions. However, garbled circuits require excessive communication between server and client, impose significant storage overheads, and incur large runtime penalties. To reduce these costs, we propose an alternative to garbled circuits: Tabula, an algorithm based on secure lookup tables. Our approach precomputes lookup tables during an offline phase that contains the result of all possible nonlinear function …

arxiv client communication computation computing cs.ai cs.cr functions large network neural network penalties runtime server storage