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

Adversarial robustness has emerged as an important topic in deep learning ascarefully crafted attack samples can significantly disturb the performance of amodel. Many recent methods have proposed to improve adversarial robustness byutilizing adversarial training or model distillation, which adds additionalprocedures to model training. In this paper, we propose a new training paradigmcalled Guided Complement Entropy (GCE) that is capable of achieving"adversarial defense for free," which involves no additional procedures in theprocess of improving adversarial robustness. In addition to maximizing modelprobabilities on the ground-truth class like cross-entropy, we neutralize itsprobabilities on the incorrect classes along with a "guided" term to balancebetween these two terms. We show in the experiments that our method achievesbetter model robustness with even better performance compared to the commonlyused cross-entropy training objective. We also show that our method can be usedorthogonal to adversarial training across well-known methods with noticeablerobustness gain. To the best of our knowledge, our approach is the first onethat improves model robustness without compromising performance.

Improving Adversarial Robustness via Guided Complement Entropy