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

Data augmentation have been intensively used in training deep neural networkto improve the generalization, whether in original space (e.g., image space) orrepresentation space. Although being successful, the connection between thesynthesized data and the original data is largely ignored in training, withoutconsidering the distribution information that the synthesized samples aresurrounding the original sample in training. Hence, the behavior of the networkis not optimized for this. However, that behavior is crucially important forgeneralization, even in the adversarial setting, for the safety of the deeplearning system. In this work, we propose a framework called Stochastic BatchAugmentation (SBA) to address these problems. SBA stochastically decideswhether to augment at iterations controlled by the batch scheduler and in whicha ''distilled'' dynamic soft label regularization is introduced byincorporating the similarity in the vicinity distribution respect to rawsamples. The proposed regularization provides direct supervision by theKL-Divergence between the output soft-max distributions of original and virtualdata. Our experiments on CIFAR-10, CIFAR-100, and ImageNet show that SBA canimprove the generalization of the neural networks and speed up the convergenceof network training.

Stochastic Batch Augmentation with An Effective Distilled Dynamic Soft Label Regularizer