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) is a powerful family of models thatlearn an underlying distribution to generate synthetic data. Many existingstudies of GANs focus on improving the realness of the generated image data forvisual applications, and few of them concern about improving the quality of thegenerated data for training other classifiers -- a task known as the modelcompatibility problem. As a consequence, existing GANs often prefer generating`easier' synthetic data that are far from the boundaries of the classifiers,and refrain from generating near-boundary data, which are known to play animportant roles in training the classifiers. To improve GAN in terms of modelcompatibility, we propose Boundary-Calibration GANs (BCGANs), which leveragethe boundary information from a set of pre-trained classifiers using theoriginal data. In particular, we introduce an auxiliary Boundary-Calibrationloss (BC-loss) into the generator of GAN to match the statistics between theposterior distributions of original data and generated data with respect to theboundaries of the pre-trained classifiers. The BC-loss is provably unbiased andcan be easily coupled with different GAN variants to improve their modelcompatibility. Experimental results demonstrate that BCGANs not only generaterealistic images like original GANs but also achieves superior modelcompatibility than the original GANs.

Improving Model Compatibility of Generative Adversarial Networks by Boundary Calibration