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

Many deep Convolutional Neural Networks (CNN) make incorrect predictions onadversarial samples obtained by imperceptible perturbations of clean samples.We hypothesize that this is caused by a failure to suppress unusual signalswithin network layers. As remedy we propose the use of Symmetric ActivationFunctions (SAF) in non-linear signal transducer units. These units suppresssignals of exceptional magnitude. We prove that SAF networks can performclassification tasks to arbitrary precision in a simplified situation. Inpractice, rather than use SAFs alone, we add them into CNNs to improve theirrobustness. The modified CNNs can be easily trained using popular strategieswith the moderate training load. Our experiments on MNIST and CIFAR-10 showthat the modified CNNs perform similarly to plain ones on clean samples, andare remarkably more robust against adversarial and nonsense samples.

Suppressing the Unusual: towards Robust CNNs using Symmetric Activation Functions