Veronika Cheplygina, Lauge Sørensen, David M. J. Tax, Marleen de Bruijne, Marco Loog
TL;DR本文探讨了在多实例学习分类器中的实例标签稳定性问题,提出了一个评估实例稳定性的无监督度量,并演示了在比较 MIL 分类器时可以进行性能 - 稳定性权衡。
Abstract
We address the problem of \emph{instance label stability} in multiple
instance learning (MIL) classifiers. These classifiers are trained only on
globally annotated images (bags), but often can provide fine-grained
annotations for image pixels or patches (instances). This is interesting