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

Building robust deterministic neural networks remains a challenge. On the onehand, some approaches improve out-of-distribution detection at the cost ofreducing classification accuracy in some situations. On the other hand, somemethods simultaneously increase classification accuracy, uncertaintyestimation, and out-of-distribution detection at the expense of reducing theinference efficiency. In this paper, we propose training deterministic neuralnetworks using our DisMax loss, which works as a drop-in replacement for theusual SoftMax loss (i.e., the combination of the linear output layer, theSoftMax activation, and the cross-entropy loss). Starting from the IsoMax+loss, we create each logit based on the distances to all prototypes, ratherthan just the one associated with the correct class. We also introduce amechanism to combine images to construct what we call fractional probabilityregularization. Moreover, we present a fast way to calibrate the network aftertraining. Finally, we propose a composite score to perform out-of-distributiondetection. Our experiments show that DisMax usually outperforms currentapproaches simultaneously in classification accuracy, uncertainty estimation,and out-of-distribution detection while maintaining deterministic neuralnetwork inference efficiency. The code to reproduce the results is available athttps://github.com/dlmacedo/distinction-maximization-loss.

Distinction Maximization Loss: Efficiently Improving Out-of-Distribution Detection and Uncertainty Estimation by Replacing the Loss and Calibrating