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

Although Deep Neural Networks (DNNs) achieve excellent performance on manyreal-world tasks, they are highly vulnerable to adversarial attacks. A leadingdefense against such attacks is adversarial training, a technique in which aDNN is trained to be robust to adversarial attacks by introducing adversarialnoise to its input. This procedure is effective but must be done during thetraining phase. In this work, we propose Augmented Random Forest (ARF), asimple and easy-to-use strategy for robustifying an existing pretrained DNNwithout modifying its weights. For every image, we generate randomized testtime augmentations by applying diverse color, blur, noise, and geometrictransforms. Then we use the DNN's logits output to train a simple random forestto predict the real class label. Our method achieves state-of-the-artadversarial robustness on a diversity of white and black box attacks withminimal compromise on the natural images' classification. We test ARF alsoagainst numerous adaptive white-box attacks and it shows excellent results whencombined with adversarial training. Code is available athttps://github.com/giladcohen/ARF.

Simple Post-Training Robustness Using Test Time Augmentations and Random Forest