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

While Deep Neural Networks (DNNs) achieve state-of-the-art accuracy invarious applications, they often fall short in accurately estimating theirpredictive uncertainty and, in turn, fail to recognize when these predictionsmay be wrong. Several uncertainty-aware models, such as Bayesian Neural Network(BNNs) and Deep Ensembles have been proposed in the literature for quantifyingpredictive uncertainty. However, research in this area has been largelyconfined to the big data regime. In this work, we show that the uncertaintyestimation capability of state-of-the-art BNNs and Deep Ensemble modelsdegrades significantly when the amount of training data is small. To addressthe issue of accurate uncertainty estimation in the small-data regime, wepropose a probabilistic generalization of the popular sample-efficientnon-parametric kNN approach. Our approach enables deep kNN classifier toaccurately quantify underlying uncertainties in its prediction. We demonstratethe usefulness of the proposed approach by achieving superior uncertaintyquantification as compared to state-of-the-art on a real-world application ofCOVID-19 diagnosis from chest X-Rays. Our code is available athttps://github.com/ankurmallick/sample-efficient-uq

Probabilistic Neighbourhood Component Analysis: Sample Efficient Uncertainty Estimation in Deep Learning