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

In many real-world applications of Machine Learning it is of paramountimportance not only to provide accurate predictions, but also to ensure certainlevels of robustness. Adversarial Training is a training procedure aiming atproviding models that are robust to worst-case perturbations around predefinedpoints. Unfortunately, one of the main issues in adversarial training is thatrobustness w.r.t. gradient-based attackers is always achieved at the cost ofprediction accuracy. In this paper, a new algorithm, called WassersteinProjected Gradient Descent (WPGD), for adversarial training is proposed. WPGDprovides a simple way to obtain cost-sensitive robustness, resulting in a finercontrol of the robustness-accuracy trade-off. Moreover, WPGD solves an optimaltransport problem on the output space of the network and it can efficientlydiscover directions where robustness is required, allowing to control thedirectional trade-off between accuracy and robustness. The proposed WPGD isvalidated in this work on image recognition tasks with different benchmarkdatasets and architectures. Moreover, real world-like datasets are oftenunbalanced: this paper shows that when dealing with such type of datasets, theperformance of adversarial training are mainly affected in term of standardaccuracy.

Directional Adversarial Training for Cost Sensitive Deep Learning Classification Applications