English

On Lipschitz Continuity and Smoothness of Loss Functions in Learning to Rank

Machine Learning 2016-09-14 v3 Machine Learning

Abstract

In binary classification and regression problems, it is well understood that Lipschitz continuity and smoothness of the loss function play key roles in governing generalization error bounds for empirical risk minimization algorithms. In this paper, we show how these two properties affect generalization error bounds in the learning to rank problem. The learning to rank problem involves vector valued predictions and therefore the choice of the norm with respect to which Lipschitz continuity and smoothness are defined becomes crucial. Choosing the \ell_\infty norm in our definition of Lipschitz continuity allows us to improve existing bounds. Furthermore, under smoothness assumptions, our choice enables us to prove rates that interpolate between 1/n1/\sqrt{n} and 1/n1/n rates. Application of our results to ListNet, a popular learning to rank method, gives state-of-the-art performance guarantees.

Keywords

Cite

@article{arxiv.1405.0586,
  title  = {On Lipschitz Continuity and Smoothness of Loss Functions in Learning to Rank},
  author = {Ambuj Tewari and Sougata Chaudhuri},
  journal= {arXiv preprint arXiv:1405.0586},
  year   = {2016}
}

Comments

This paper has been withdrawn as it was superseded by an ICML 2015 paper "Generalization error bounds for learning to rank: Does the length of document lists matter?" available as arXiv:1603.01860