Efficient Bayesian Inference for Generalized Bradley–Terry Models

  title={Efficient Bayesian Inference for Generalized Bradley–Terry Models},
  author={François Caron and A. Doucet},
  journal={Journal of Computational and Graphical Statistics},
  pages={174 - 196}
  • F. Caron, A. Doucet
  • Published 8 November 2010
  • Computer Science
  • Journal of Computational and Graphical Statistics
The Bradley–Terry model is a popular approach to describe probabilities of the possible outcomes when elements of a set are repeatedly compared with one another in pairs. It has found many applications including animal behavior, chess ranking, and multiclass classification. Numerous extensions of the basic model have also been proposed in the literature including models with ties, multiple comparisons, group comparisons, and random graphs. From a computational point of view, Hunter has proposed… 
A Bayesian nonparametric approach for generalized Bradley-Terry models in random environment
A Bayesian nonparametric approach to solve the estimation of the unknown distribution of hidden random variables from the observation of pairwise comparisons between these variables and a Markov Chain Monte Carlo algorithm is proposed.
Convergence Rates of Gradient Descent and MM Algorithms for Generalized Bradley-Terry Models
We show tight convergence rate bounds for gradient descent and MM algorithms for maximum likelihood estimation and maximum aposteriori probability estimation of a popular Bayesian inference method
Plackett-Luce regression: A new Bayesian model for polychotomous data
This paper builds on the Plackett-Luce model, a popular model for multiple comparisons, and shows that the introduction of a suitable set of auxiliary variables leads to an Expectation-Maximization algorithm to find Maximum A Posteriori estimates of the parameters.
Bayesian Plackett–Luce Mixture Models for Partially Ranked Data
A Bayesian finite mixture of Plackett–Luce models to account for unobserved sample heterogeneity of partially ranked data is introduced and an efficient way to incorporate the latent group structure in the data augmentation approach and the derivation of existing maximum likelihood procedures as special instances of the proposed Bayesian method is described.
Revisiting Random Utility Models
This thesis explores extensions of Random Utility Models, providing more exible models and adopting a computational perspective, and proposes two preference elicitation scheme for generalized RUMs, in which the utilities can also depend on attributes of agents and alternatives.
On Bayesian inference for the Extended Plackett-Luce model
This paper considers the Extended Plackett-Luce model that induces a flexible (discrete) distribution over permutations that draws on this distribution to address the rank aggregation problem and also to identify potential lack of model fit.
Lower Bounds on the Bayes Risk of the Bayesian BTL Model With Applications to Comparison Graphs
This work considers the problem of aggregating pairwise comparisons to obtain a consensus ranking order over a collection of objects using the popular Bradley–Terry–Luce model, and derives information-theoretic lower bounds on the Bayes risk of estimators for norm-based distortion functions.
Ranking in the generalized Bradley–Terry models when the strong connection condition fails
ABSTRACT For non balanced paired comparisons, a wide variety of ranking methods have been proposed. One of the best popular methods is the Bradley–Terry model in which the ranking of a set of objects
Improving pairwise comparison models using Empirical Bayes shrinkage
This work derives a collection of methods for estimating the pairwise uncertainty of pairwise predictions based on different assumptions about the comparison process to examine model uncertainty as well as perform Empirical Bayes shrinkage estimation of the model parameters.
Bayesian analysis of ranking data with the Extended Plackett–Luce model
This work proposes the Bayesian estimation of the EPL in order to address more directly and efficiently the inference on the additional discrete-valued parameter and the assessment of its estimation uncertainty, possibly uncovering potential idiosyncratic drivers in the formation of preferences.


MM algorithms for generalized Bradley-Terry models
The Bradley-Terry model for paired comparisons is a simple and muchstudied means to describe the probabilities of the possible outcomes when individuals are judged against one another in pairs. Among
Generalized Bradley-Terry Models and Multi-Class Probability Estimates
This paper introduces a generalized Bradley-Terry model in which paired individual comparisons are extended to paired team comparisons, and proposes a simple algorithm with convergence proofs to solve the model and obtain individual skill.
Bayesian inference for Plackett-Luce ranking models
An efficient Bayesian method for inferring the parameters of a Plackett-Luce ranking model is given and a number of advantages of the EP approach over the traditional maximum likelihood method are shown.
A choice model with infinitely many latent features
An MCMC algorithm for the EBA model is presented that can also be used in inference for other non-conjugate IBP models---this may broaden the use of IBP priors considerably.
A Matlab function to estimate choice model parameters from paired-comparison data
A Matlab function is presented that makes it easy to specify any of these general models for paired-comparison data (EBA, Pretree, or BTL) and to estimate their parameters and eliminates the time-consuming task of constructing the likelihood function by hand for every single model.
Bayesian analysis of linear dominance hierarchies
Ties in Paired-Comparison Experiments: A Generalization of the Bradley-Terry Model
Abstract The Bradley-Terry model for a paired-comparison experiment with t treatments postulates a set of t ‘true’ treatment ratings π1, π2, · · ·, π t such that π i ≥ 0, ∑ π i = 1 and the
Gibbs sampling for Bayesian non‐conjugate and hierarchical models by using auxiliary variables
The aim of the paper is to provide an alternative sampling algorithm to rejection‐based methods and other sampling approaches such as the Metropolis–Hastings algorithm.
Random graphs with a given degree sequence
Large graphs are sometimes studied through their degree sequences (power law or regular graphs). We study graphs that are uniformly chosen with a given degree sequence. Under mild conditions, it is
An Exponential Family of Probability Distributions for Directed Graphs
An exponential family of distributions that can be used for analyzing directed graph data is described, and several special cases are discussed along with some possible substantive interpretations.