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

Recent work has shown that deep generative models assign higher likelihood toout-of-distribution inputs than to training data. We show that a factorunderlying this phenomenon is a mismatch between the nature of the priordistribution and that of the data distribution, a problem found in widely useddeep generative models such as VAEs and Glow. While a typical choice for aprior distribution is a standard Gaussian distribution, properties ofdistributions of real data sets may not be consistent with a unimodal priordistribution. This paper focuses on the relationship between the choice of aprior distribution and the likelihoods assigned to out-of-distribution inputs.We propose the use of a mixture distribution as a prior to make likelihoodsassigned by deep generative models sensitive to out-of-distribution inputs.Furthermore, we explain the theoretical advantages of adopting a mixturedistribution as the prior, and we present experimental results to support ourclaims. Finally, we demonstrate that a mixture prior lowers theout-of-distribution likelihood with respect to two pairs of real image datasets: Fashion-MNIST vs. MNIST and CIFAR10 vs. SVHN.

Likelihood Assignment for Out-of-Distribution Inputs in Deep Generative Models is Sensitive to Prior Distribution Choice