English

Learning-to-Rank with Partitioned Preference: Fast Estimation for the Plackett-Luce Model

Machine Learning 2021-03-01 v3 Machine Learning

Abstract

We investigate the Plackett-Luce (PL) model based listwise learning-to-rank (LTR) on data with partitioned preference, where a set of items are sliced into ordered and disjoint partitions, but the ranking of items within a partition is unknown. Given NN items with MM partitions, calculating the likelihood of data with partitioned preference under the PL model has a time complexity of O(N+S!)O(N+S!), where SS is the maximum size of the top M1M-1 partitions. This computational challenge restrains most existing PL-based listwise LTR methods to a special case of partitioned preference, top-KK ranking, where the exact order of the top KK items is known. In this paper, we exploit a random utility model formulation of the PL model, and propose an efficient numerical integration approach for calculating the likelihood and its gradients with a time complexity O(N+S3)O(N+S^3). We demonstrate that the proposed method outperforms well-known LTR baselines and remains scalable through both simulation experiments and applications to real-world eXtreme Multi-Label classification tasks.

Keywords

Cite

@article{arxiv.2006.05067,
  title  = {Learning-to-Rank with Partitioned Preference: Fast Estimation for the Plackett-Luce Model},
  author = {Jiaqi Ma and Xinyang Yi and Weijing Tang and Zhe Zhao and Lichan Hong and Ed H. Chi and Qiaozhu Mei},
  journal= {arXiv preprint arXiv:2006.05067},
  year   = {2021}
}
R2 v1 2026-06-23T16:10:09.212Z