Gradient-free Gradient Boosting family for the normalized weak ranking loss function.
WeakRankNorm(K)Indicates that we are only interesting in the top \(K\) instances. Must be an integer between 1 and the number \(n\) of observations.
A Boosting family object
A more intuitive loss function than the weak ranking loss thanks to its normalization to a maximum value
of 1. For example, if a number \(c\) of the top \(K\) instances has not been ranked at the top of the list, the
normalized weak ranking loss is \(C/K\). WeakRankNorm returns a family object as in the package mboost.
Werner, T., Gradient-Free Gradient Boosting, PhD Thesis, Carl von Ossietzky University Oldenburg, 2020, Remark (5.2.4)
T. Hothorn, P. B<U+00FC>hlmann, T. Kneib, M. Schmid, and B. Hofner. mboost: Model-Based Boosting, 2017