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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The large number of ReLU non-linearity operations in existing deep neuralnetworks makes them ill-suited for latency-efficient private inference (PI).Existing techniques to reduce ReLU operations often involve manual effort andsacrifice significant accuracy. In this paper, we first present a novel measureof non-linearity layers' ReLU sensitivity, enabling mitigation of thetime-consuming manual efforts in identifying the same. Based on thissensitivity, we then present SENet, a three-stage training method that for agiven ReLU budget, automatically assigns per-layer ReLU counts, decides theReLU locations for each layer's activation map, and trains a model withsignificantly fewer ReLUs to potentially yield latency and communicationefficient PI. Experimental evaluations with multiple models on various datasetsshow SENet's superior performance both in terms of reduced ReLUs and improvedclassification accuracy compared to existing alternatives. In particular, SENetcan yield models that require up to ~2x fewer ReLUs while yielding similaraccuracy. For a similar ReLU budget SENet can yield models with ~2.32% improvedclassification accuracy, evaluated on CIFAR-100.

Learning to Linearize Deep Neural Networks for Secure and Efficient Private Inference