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 consider the problem of uncertainty estimation in the context of(non-Bayesian) deep neural classification. In this context, all known methodsare based on extracting uncertainty signals from a trained network optimized tosolve the classification problem at hand. We demonstrate that such techniquestend to introduce biased estimates for instances whose predictions are supposedto be highly confident. We argue that this deficiency is an artifact of thedynamics of training with SGD-like optimizers, and it has some propertiessimilar to overfitting. Based on this observation, we develop an uncertaintyestimation algorithm that selectively estimates the uncertainty of highlyconfident points, using earlier snapshots of the trained model, before theirestimates are jittered (and way before they are ready for actualclassification). We present extensive experiments indicating that the proposedalgorithm provides uncertainty estimates that are consistently better than allknown methods.

Bias-Reduced Uncertainty Estimation for Deep Neural Classifiers