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

Dealing with label noise in tabular classification tasks poses a persistentchallenge in machine learning. While robust boosting methods have shown promisein binary classification, their effectiveness in complex, multi-class scenariosis often limited. Additionally, issues like imbalanced datasets, missingvalues, and computational inefficiencies further complicate their practicalutility. This study introduces Robust-GBDT, a groundbreaking approach thatcombines the power of Gradient Boosted Decision Trees (GBDT) with theresilience of nonconvex loss functions against label noise. By leveraging localconvexity within specific regions, Robust-GBDT demonstrates unprecedentedrobustness, challenging conventional wisdom. Through seamless integration ofadvanced GBDT with a novel Robust Focal Loss tailored for class imbalance,Robust-GBDT significantly enhances generalization capabilities, particularly innoisy and imbalanced datasets. Notably, its user-friendly design facilitatesintegration with existing open-source code, enhancing computational efficiencyand scalability. Extensive experiments validate Robust-GBDT's superiority overother noise-robust methods, establishing a new standard for accurateclassification amidst label noise. This research heralds a paradigm shift inmachine learning, paving the way for a new era of robust and preciseclassification across diverse real-world applications.

Robust-GBDT: GBDT with Nonconvex Loss for Tabular Classification in the Presence of Label Noise and Class Imbalance