General Oracle Inequalities for Gibbs Posterior with Application to Ranking

Abstract

In this paper, we summarize some recent results in Li et al. (2012), which can be used to extend an important PAC-Bayesian approach, namely the Gibbs posterior, to study the nonadditive ranking risk. The methodology is based on assumption-free risk bounds and nonasymptotic oracle inequalities, which leads to nearly optimal convergence rates and optimal model selection to balance the approximation errors and the stochastic errors.

Cite

Text

Li et al. "General Oracle Inequalities for Gibbs Posterior with Application to Ranking." Annual Conference on Computational Learning Theory, 2013.

Markdown

[Li et al. "General Oracle Inequalities for Gibbs Posterior with Application to Ranking." Annual Conference on Computational Learning Theory, 2013.](https://mlanthology.org/colt/2013/li2013colt-general/)

BibTeX

@inproceedings{li2013colt-general,
  title     = {{General Oracle Inequalities for Gibbs Posterior with Application to Ranking}},
  author    = {Li, Cheng and Jiang, Wenxin and Tanner, Martin A.},
  booktitle = {Annual Conference on Computational Learning Theory},
  year      = {2013},
  pages     = {512-521},
  url       = {https://mlanthology.org/colt/2013/li2013colt-general/}
}