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

Data augmentation has been widely used to improve generalization in trainingdeep neural networks. Recent works show that using worst-case transformationsor adversarial augmentation strategies can significantly improve the accuracyand robustness. However, due to the non-differentiable properties of imagetransformations, searching algorithms such as reinforcement learning orevolution strategy have to be applied, which are not computationally practicalfor large scale problems. In this work, we show that by simply applyingconsistency training with random data augmentation, state-of-the-art results ondomain adaptation (DA) and generalization (DG) can be obtained. To furtherimprove the accuracy and robustness with adversarial examples, we propose adifferentiable adversarial data augmentation method based on spatialtransformer networks (STN). The combined adversarial and random transformationsbased method outperforms the state-of-the-art on multiple DA and DG benchmarkdatasets. Besides, the proposed method shows desirable robustness tocorruption, which is also validated on commonly used datasets.

Adversarial and Random Transformations for Robust Domain Adaptation and Generalization