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

Multi-label learning in the presence of missing labels (MLML) is achallenging problem. Existing methods mainly focus on the design of networkstructures or training schemes, which increase the complexity ofimplementation. This work seeks to fulfill the potential of loss function inMLML without increasing the procedure and complexity. Toward this end, wepropose two simple yet effective methods via robust loss design based on anobservation that a model can identify missing labels during training with ahigh precision. The first is a novel robust loss for negatives, namely the Hillloss, which re-weights negatives in the shape of a hill to alleviate the effectof false negatives. The second is a self-paced loss correction (SPLC) method,which uses a loss derived from the maximum likelihood criterion under anapproximate distribution of missing labels. Comprehensive experiments on a vastrange of multi-label image classification datasets demonstrate that our methodscan remarkably boost the performance of MLML and achieve new state-of-the-artloss functions in MLML.

Simple and Robust Loss Design for Multi-Label Learning with Missing Labels