Composite Multiclass Losses
We consider loss functions for multiclass prediction problems. We show when a multiclass loss can be expressed as a "proper composite loss", which is the composition of a proper loss and a link function. We extend existing results for binary losses to mul
|Collections||ANU Research Publications|
|Source:||Advances in Neural Information Processing Systems 23|
|01_Vernet_Composite_Multiclass_Los_2011.pdf||469.95 kB||Adobe PDF|
Items in Open Research are protected by copyright, with all rights reserved, unless otherwise indicated.