On Generalizing the C-Bound to the Multiclass and Multi-label Settings
Machine Learning
2015-01-14 v1 Machine Learning
Abstract
The C-bound, introduced in Lacasse et al., gives a tight upper bound on the risk of a binary majority vote classifier. In this work, we present a first step towards extending this work to more complex outputs, by providing generalizations of the C-bound to the multiclass and multi-label settings.
Cite
@article{arxiv.1501.03001,
title = {On Generalizing the C-Bound to the Multiclass and Multi-label Settings},
author = {Francois Laviolette and Emilie Morvant and Liva Ralaivola and Jean-Francis Roy},
journal= {arXiv preprint arXiv:1501.03001},
year = {2015}
}
Comments
NIPS 2014 Workshop on Representation and Learning Methods for Complex Outputs, Dec 2014, Montr{\'e}al, Canada