Recognition: unknown
A more robust boosting algorithm
classification
📊 stat.ML
keywords
algorithmboostingrobustevidenceexistingexperimentalgivelabel
read the original abstract
We present a new boosting algorithm, motivated by the large margins theory for boosting. We give experimental evidence that the new algorithm is significantly more robust against label noise than existing boosting algorithm.
This paper has not been read by Pith yet.
Forward citations
Cited by 2 Pith papers
-
ITBoost: Information-Theoretic Trust for Robust Boosting
ITBoost applies MDL-based trust scoring on residual histories to down-weight unreliable samples in gradient boosting, claiming a tighter generalization bound and improved robustness to label noise on tabular tasks.
-
ITBoost: Information-Theoretic Trust for Robust Boosting
ITBoost uses MDL-based complexity of residual trajectories to assign trust weights, improving robustness to label noise in tabular boosting without sacrificing clean-data performance.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.