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

We introduce Exemplar VAEs, a family of generative models that bridge the gapbetween parametric and non-parametric, exemplar based generative models.Exemplar VAE is a variant of VAE with a non-parametric prior in the latentspace based on a Parzen window estimator. To sample from it, one first draws arandom exemplar from a training set, then stochastically transforms thatexemplar into a latent code and a new observation. We propose retrievalaugmented training (RAT) as a way to speed up Exemplar VAE training by usingapproximate nearest neighbor search in the latent space to define a lower boundon log marginal likelihood. To enhance generalization, model parameters arelearned using exemplar leave-one-out and subsampling. Experiments demonstratethe effectiveness of Exemplar VAEs on density estimation and representationlearning. Importantly, generative data augmentation using Exemplar VAEs onpermutation invariant MNIST and Fashion MNIST reduces classification error from1.17% to 0.69% and from 8.56% to 8.16%.

Exemplar VAE: Linking Generative Models, Nearest Neighbor Retrieval, and Data Augmentation