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

Binary Neural Networks (BNNs), neural networks with weights and activationsconstrained to -1(0) and +1, are an alternative to deep neural networks whichoffer faster training, lower memory consumption and lightweight models, idealfor use in resource constrained devices while being able to utilize thearchitecture of their deep neural network counterpart. However, the inputbinarization step used in BNNs causes a severe accuracy loss. In this paper, weintroduce a novel hybrid neural network architecture, Hybrid Binary NeuralNetwork (HyBNN), consisting of a task-independent, general, full-precisionvariational autoencoder with a binary latent space and a task specific binaryneural network that is able to greatly limit the accuracy loss due to inputbinarization by using the full precision variational autoencoder as a featureextractor. We use it to combine the state-of-the-art accuracy of deep neuralnetworks with the much faster training time, quicker test-time inference andpower efficiency of binary neural networks. We show that our proposed system isable to very significantly outperform a vanilla binary neural network withinput binarization. We also introduce FedHyBNN, a highly communicationefficient federated counterpart to HyBNN and demonstrate that it is able toreach the same accuracy as its non-federated equivalent. We make our sourcecode, experimental parameters and models available at:https://anonymous.4open.science/r/HyBNN.

HyBNN and FedHyBNN: (Federated) Hybrid Binary Neural Networks