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/

Zendy Plus

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

Zendy Tools

Zendy Open

During training, the weights of a Deep Neural Network (DNN) are optimizedfrom a random initialization towards a nearly optimum value minimizing a lossfunction. Only this final state of the weights is typically kept for testing,while the wealth of information on the geometry of the weight space,accumulated over the descent towards the minimum is discarded. In this work wepropose to make use of this knowledge and leverage it for computing thedistributions of the weights of the DNN. This can be further used forestimating the epistemic uncertainty of the DNN by sampling an ensemble ofnetworks from these distributions. To this end we introduce a method fortracking the trajectory of the weights during optimization, that does notrequire any changes in the architecture nor on the training procedure. Weevaluate our method on standard classification and regression benchmarks, andon out-of-distribution detection for classification and semantic segmentation.We achieve competitive results, while preserving computational efficiency incomparison to other popular approaches.

TRADI: Tracking deep neural network weight distributions for uncertainty estimation