On the identifiability of mixtures of ranking models
Preprint
Abstract
Mixtures of ranking models are standard tools for ranking problems. However, even the fundamental question of parameter identifiability is not fully understood: the identifiability of a mixture model with two Bradley-Terry-Luce (BTL) components has remained open. In this work, we show that popular mixtures of ranking models with two components (BTL, multinomial logistic models with slates of size 3, or Plackett-Luce) are generically identifiable, i.e., the ground-truth parameters can be identified except when they are from a pathological subset of measure zero. We provide a framework for verifying the number of solutions in a general family of polynomial systems using algebraic geometry, and apply it to these mixtures of ranking models to establish generic identifiability. The framework can be applied more broadly to other learning models and may be of independent interest.Keywords
All Related Versions
This publication has 10 references indexed in Scilit:
- Learning mixtures of permutations: Groups of pairwise comparisons and combinatorial method of momentsThe Annals of Statistics, 2022
- Moment Identifiability of Homoscedastic Gaussian MixturesFoundations of Computational Mathematics, 2020
- Algebraic Identifiability of Gaussian MixturesInternational Mathematics Research Notices, 2017
- Identifiability of 3-class Jukes–Cantor mixturesAdvances in Applied Mathematics, 2015
- Commutative algebra of statistical rankingJournal of Algebra, 2012
- On the identifiability of a mixture model for ordinal dataMETRON, 2010
- Lectures on Algebraic Statistics2009
- The Analysis of PermutationsJournal of the Royal Statistical Society Series C: Applied Statistics, 1975
- Distinctness of the Eigenvalues of a Quadratic form in a Multivariate SampleThe Annals of Statistics, 1973
- Analytic Functions of Several Complex Variables.The American Mathematical Monthly, 1968