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 moment matching network (GMMN) is a deep generative model thatdiffers from Generative Adversarial Network (GAN) by replacing thediscriminator in GAN with a two-sample test based on kernel maximum meandiscrepancy (MMD). Although some theoretical guarantees of MMD have beenstudied, the empirical performance of GMMN is still not as competitive as thatof GAN on challenging and large benchmark datasets. The computationalefficiency of GMMN is also less desirable in comparison with GAN, partially dueto its requirement for a rather large batch size during the training. In thispaper, we propose to improve both the model expressiveness of GMMN and itscomputational efficiency by introducing adversarial kernel learning techniques,as the replacement of a fixed Gaussian kernel in the original GMMN. The newapproach combines the key ideas in both GMMN and GAN, hence we name it MMD GAN.The new distance measure in MMD GAN is a meaningful loss that enjoys theadvantage of weak topology and can be optimized via gradient descent withrelatively small batch sizes. In our evaluation on multiple benchmark datasets,including MNIST, CIFAR- 10, CelebA and LSUN, the performance of MMD-GANsignificantly outperforms GMMN, and is competitive with other representativeGAN works.

MMD GAN: Towards Deeper Understanding of Moment Matching Network