Conditional independence and graphical models for rankings
Abstract
Ranking models define probability distributions over rankings and are fundamental in applications such as preference learning and decision-making. Without structural assumptions, these models are intractable due to their factorial complexity. Although independence assumptions are key to reducing complexity and enabling interpretability, standard notions of conditional independence must account for the mutual exclusivity constraints inherent to rankings. We introduce relative rank models, a probabilistic framework based on structured coarsenings of the ranking space that relax these constraints while preserving relative order information. This representation enables a well-defined notion of conditional independence for rankings and allows rankings to be modeled using Bayesian networks. As a result, standard methods for learning and probabilistic inference can be applied to rankings, with interpretable representations based on conditional independence. We further show that two prominent ranking models based on independence assumptions tailored to rankings arise as special cases, providing a unifying framework for independence in ranking models.
Related
- A Kernel-Based Test for Multi-Way Conditional Independence
- Sequential Conditional Independence Testing with Machine Learning Models
- Causal Discovery beyond Conditional Independence: Identifiability, Algorithms, and Statistical Inference for Directed Acyclic and Cyclic Graphs
- On the Relaxation of Conditional Independence Assumption for Image Segmentation
- Graph-Based Testing of Conditional Mean Independence with Applications to Large Language Model Evaluation
BibTeX
@misc{perez2026conditional,
title = {{Conditional independence and graphical models for rankings}},
author = {Aritz Pérez},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.04586},
eprint = {2611.04586},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.04586},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Conditional independence and graphical models for rankings
AU - Aritz Pérez
PY - 2026
DA - 2026/11/03
PB - arXiv
DO - 10.48550/arxiv.2611.04586
UR - https://arxiv.org/abs/2611.04586
AB - Ranking models define probability distributions over rankings and are fundamental in applications such as preference learning and decision-making. Without structural assumptions, these models are intractable due to their factorial complexity. Although independence assumptions are key to reducing complexity and enabling interpretability, standard notions of conditional independence must account for the mutual exclusivity constraints inherent to rankings. We introduce relative rank models, a probabilistic framework based on structured coarsenings of the ranking space that relax these constraints while preserving relative order information. This representation enables a well-defined notion of conditional independence for rankings and allows rankings to be modeled using Bayesian networks. As a result, standard methods for learning and probabilistic inference can be applied to rankings, with interpretable representations based on conditional independence. We further show that two prominent ranking models based on independence assumptions tailored to rankings arise as special cases, providing a unifying framework for independence in ranking models.
ER -