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

Imbalanced data pose challenges for deep learning based classificationmodels. One of the most widely-used approaches for tackling imbalanced data isre-weighting, where training samples are associated with different weights inthe loss function. Most of existing re-weighting approaches treat the exampleweights as the learnable parameter and optimize the weights on the meta set,entailing expensive bilevel optimization. In this paper, we propose a novelre-weighting method based on optimal transport (OT) from a distributional pointof view. Specifically, we view the training set as an imbalanced distributionover its samples, which is transported by OT to a balanced distributionobtained from the meta set. The weights of the training samples are theprobability mass of the imbalanced distribution and learned by minimizing theOT distance between the two distributions. Compared with existing methods, ourproposed one disengages the dependence of the weight learning on the concernedclassifier at each iteration. Experiments on image, text and point clouddatasets demonstrate that our proposed re-weighting method has excellentperformance, achieving state-of-the-art results in many cases and providing apromising tool for addressing the imbalanced classification issue.

Learning to Re-weight Examples with Optimal Transport for Imbalanced Classification