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

Generative Adversarial Networks (GANs) based semi-supervised learning (SSL)approaches are shown to improve classification performance by utilizing a largenumber of unlabeled samples in conjunction with limited labeled samples.However, their performance still lags behind the state-of-the-art non-GAN basedSSL approaches. We identify that the main reason for this is the lack ofconsistency in class probability predictions on the same image under localperturbations. Following the general literature, we address this issue vialabel consistency regularization, which enforces the class probabilitypredictions for an input image to be unchanged under varioussemantic-preserving perturbations. In this work, we introduce consistencyregularization into the vanilla semi-GAN to address this critical limitation.In particular, we present a new composite consistency regularization methodwhich, in spirit, leverages both local consistency and interpolationconsistency. We demonstrate the efficacy of our approach on two SSL imageclassification benchmark datasets, SVHN and CIFAR-10. Our experiments show thatthis new composite consistency regularization based semi-GAN significantlyimproves its performance and achieves new state-of-the-art performance amongGAN-based SSL approaches.

Consistency Regularization with Generative Adversarial Networks for Semi-Supervised Learning