English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

While it is encouraging to witness the recent development inprivacy-preserving Machine Learning as a Service (MLaaS), there still exists asignificant performance gap for its deployment in real-world applications. Weobserve the state-of-the-art frameworks follow a compute-and-share principlefor every function output where the summing in linear functions, which is thelast of two steps for function output, involves all rotations (which is themost expensive HE operation), and the multiplexing in nonlinear functions,which is also the last of two steps for function output, introduces noticeablecommunication rounds. Therefore, we challenge the conventionalcompute-and-share logic and introduce the first joint linear and nonlinearcomputation across functions that features by 1) the PHE triplet for computingthe nonlinear function, with which the multiplexing is eliminated; 2) thematrix encoding to calculate the linear function, with which all rotations forsumming is removed; and 3) the network adaptation to reassemble the modelstructure, with which the joint computation module is utilized as much aspossible. The boosted efficiency is verified by the numerical complexity, andthe experiments demonstrate up to 13x speedup for various functions used in thestate-of-the-art models and up to 5x speedup over mainstream neural networks.

Joint Linear and Nonlinear Computation across Functions for Efficient Privacy-Preserving Neural Network Inference