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

Deep learning has revolutionized the performance of classification, butmeanwhile demands sufficient labeled data for training. Given insufficientdata, while many techniques have been developed to help combat overfitting, thechallenge remains if one tries to train deep networks, especially in theill-posed extremely low data regimes: only a small set of labeled data areavailable, and nothing -- including unlabeled data -- else. Such regimes arisefrom practical situations where not only data labeling but also data collectionitself is expensive. We propose a deep adversarial data augmentation (DADA)technique to address the problem, in which we elaborately formulate dataaugmentation as a problem of training a class-conditional and supervisedgenerative adversarial network (GAN). Specifically, a new discriminator loss isproposed to fit the goal of data augmentation, through which both real andaugmented samples are enforced to contribute to and be consistent in findingthe decision boundaries. Tailored training techniques are developedaccordingly. To quantitatively validate its effectiveness, we first performextensive simulations to show that DADA substantially outperforms bothtraditional data augmentation and a few GAN-based options. We then extendexperiments to three real-world small labeled datasets where existing dataaugmentation and/or transfer learning strategies are either less effective orinfeasible. All results endorse the superior capability of DADA in enhancingthe generalization ability of deep networks trained in practical extremely lowdata regimes. Source code is available athttps://github.com/SchafferZhang/DADA.

DADA: Deep Adversarial Data Augmentation for Extremely Low Data Regime Classification