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

Deep neural networks have recently achieved state-of-the-art results in manymachine learning problems, e.g., speech recognition or object recognition.Hitherto, work on rectified linear units (ReLU) provides empirical andtheoretical evidence on performance increase of neural networks comparing totypically used sigmoid activation function. In this paper, we investigate a newmanner of improving neural networks by introducing a bunch of copies of thesame neuron modeled by the generalized Kumaraswamy distribution. As a result,we propose novel non-linear activation function which we refer to asKumaraswamy unit which is closely related to ReLU. In the experimental studywith MNIST image corpora we evaluate the Kumaraswamy unit applied tosingle-layer (shallow) neural network and report a significant drop in testclassification error and test cross-entropy in comparison to sigmoid unit, ReLUand Noisy ReLU.

Improving neural networks with bunches of neurons modeled by Kumaraswamy units: Preliminary study