ML Anthology

Bayesian Estimators as Voting Rules

Lirong Xia
UAI 2016
/uai/2016/xia2016uai-bayesian/

Abstract

We investigate the fairness of Bayesian estimators (BEs) by viewing them as (irresolute) voting rules and evaluating them by satisfaction of desirable social choice axioms. We characterize the class of BEs that satisfy neutrality by the class of BEs with neutral structures. We prove that a BE with a neutral structure is a minimax rule if it further satisfies parameter connectivity. We prove that no BE satisfies strict Condorcet criterion. We also propose three new BEs of natural frameworks and investigate their computational complexity and satisfaction of monotonicity and Condorcet criterion.

PDF UAI Semantic Scholar

Cite

Text

Xia. "Bayesian Estimators as Voting Rules." Conference on Uncertainty in Artificial Intelligence, 2016.

Markdown

[Xia. "Bayesian Estimators as Voting Rules." Conference on Uncertainty in Artificial Intelligence, 2016.](https://mlanthology.org/uai/2016/xia2016uai-bayesian/)

BibTeX

@inproceedings{xia2016uai-bayesian,
  title     = {{Bayesian Estimators as Voting Rules}},
  author    = {Xia, Lirong},
  booktitle = {Conference on Uncertainty in Artificial Intelligence},
  year      = {2016},
  url       = {https://mlanthology.org/uai/2016/xia2016uai-bayesian/}
}