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

Recent work has uncovered the interesting (and somewhat surprising) findingthat training models to be invariant to adversarial perturbations requiressubstantially larger datasets than those required for standard classification.This result is a key hurdle in the deployment of robust machine learning modelsin many real world applications where labeled data is expensive. Our maininsight is that unlabeled data can be a competitive alternative to labeled datafor training adversarially robust models. Theoretically, we show that in asimple statistical setting, the sample complexity for learning an adversariallyrobust model from unlabeled data matches the fully supervised case up toconstant factors. On standard datasets like CIFAR-10, a simple UnsupervisedAdversarial Training (UAT) approach using unlabeled data improves robustaccuracy by 21.7% over using 4K supervised examples alone, and captures over95% of the improvement from the same number of labeled examples. Finally, wereport an improvement of 4% over the previous state-of-the-art on CIFAR-10against the strongest known attack by using additional unlabeled data from theuncurated 80 Million Tiny Images dataset. This demonstrates that our findingextends as well to the more realistic case where unlabeled data is alsouncurated, therefore opening a new avenue for improving adversarial training.

Are Labels Required for Improving Adversarial Robustness?