Tag
This paper resolves the minimax rate for agnostic multiclass transductive learning, demonstrating that the optimal excess error is governed by the DS and Natarajan dimensions. The result holds for arbitrary label spaces, extending prior work on binary classification.
This paper demonstrates that monotone adversarial corruptions can make certain multiclass and partial binary classification problems unlearnable, providing tight bounds on corruption budgets and extending previous results on binary classification.
This paper proposes LFS-FRAME, a leakage-free stacked ensemble framework integrating Kolmogorov-Arnold Networks and XGBoost for robust multiclass classification, achieving 89.85% accuracy on major families and 81.74% on sub-families.
This paper proposes a generalized distribution-free semi-supervised learning framework that constructs unbiased risk estimators via linear combinations of component risks, extending PNU learning to multiclass classification while achieving lower variance and providing generalization bounds.
This paper introduces the Level-Constrained-Littlestone-Littlestone (LCLL) tree to characterize learnability in universal transductive online classification with possibly unbounded label spaces, proving that optimal mistake rates are either bounded or logarithmic.
This paper characterizes approximate property calibration for discrete properties in multiclass classification, using Lipschitz continuous properties as an intermediary to reduce complexity from the number of classes to the elicitation complexity dimension.