Unified risk analysis for weakly supervised learning
CK Chiang, M Sugiyama - arxiv preprint arxiv:2309.08216, 2023 - arxiv.org
Among the flourishing research of weakly supervised learning (WSL), we recognize the lack
of a unified interpretation of the mechanism behind the weakly supervised scenarios, let …
of a unified interpretation of the mechanism behind the weakly supervised scenarios, let …
Learning with proper partial labels
Partial-label learning is a kind of weakly supervised learning with inexact labels, where for
each training example, we are given a set of candidate labels instead of only one true label …
each training example, we are given a set of candidate labels instead of only one true label …
Pairwise Confidence Difference on Unlabeled Data is Sufficient for Binary Classification
Learning with confidence labels is an emerging weakly supervised learning paradigm,
where training data are equipped with confidence labels instead of exact labels. Positive …
where training data are equipped with confidence labels instead of exact labels. Positive …
[PDF][PDF] Binary Classification from Uncertainty and Triplet Comparison
Z Cui - 2021 - repository.dl.itc.u-tokyo.ac.jp
Machine learning has achieved tremendous development and brought innovation to many
aspects of the society in the recent decade. Binary classification, one of the core tasks of …
aspects of the society in the recent decade. Binary classification, one of the core tasks of …