Preference Modeling with Context-Dependent Salient Features

Amanda Bower, Laura Balzano
Proceedings of the 37th International Conference on Machine Learning, PMLR 119:1067-1077, 2020.

Abstract

We consider the problem of estimating a ranking on a set of items from noisy pairwise comparisons given item features. We address the fact that pairwise comparison data often reflects irrational choice, e.g. intransitivity. Our key observation is that two items compared in isolation from other items may be compared based on only a salient subset of features. Formalizing this framework, we propose the salient feature preference model and prove a finite sample complexity result for learning the parameters of our model and the underlying ranking with maximum likelihood estimation. We also provide empirical results that support our theoretical bounds and illustrate how our model explains systematic intransitivity. Finally we demonstrate strong performance of maximum likelihood estimation of our model on both synthetic data and two real data sets: the UT Zappos50K data set and comparison data about the compactness of legislative districts in the US.

Cite this Paper

BibTeX
@InProceedings{pmlr-v119-bower20a, title = {Preference Modeling with Context-Dependent Salient Features}, author = {Bower, Amanda and Balzano, Laura}, booktitle = {Proceedings of the 37th International Conference on Machine Learning}, pages = {1067--1077}, year = {2020}, editor = {III, Hal Daumé and Singh, Aarti}, volume = {119}, series = {Proceedings of Machine Learning Research}, month = {13--18 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v119/bower20a/bower20a.pdf}, url = {https://proceedings.mlr.press/v119/bower20a.html}, abstract = {We consider the problem of estimating a ranking on a set of items from noisy pairwise comparisons given item features. We address the fact that pairwise comparison data often reflects irrational choice, e.g. intransitivity. Our key observation is that two items compared in isolation from other items may be compared based on only a salient subset of features. Formalizing this framework, we propose the salient feature preference model and prove a finite sample complexity result for learning the parameters of our model and the underlying ranking with maximum likelihood estimation. We also provide empirical results that support our theoretical bounds and illustrate how our model explains systematic intransitivity. Finally we demonstrate strong performance of maximum likelihood estimation of our model on both synthetic data and two real data sets: the UT Zappos50K data set and comparison data about the compactness of legislative districts in the US.} }
Endnote
%0 Conference Paper %T Preference Modeling with Context-Dependent Salient Features %A Amanda Bower %A Laura Balzano %B Proceedings of the 37th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2020 %E Hal Daumé III %E Aarti Singh %F pmlr-v119-bower20a %I PMLR %P 1067--1077 %U https://proceedings.mlr.press/v119/bower20a.html %V 119 %X We consider the problem of estimating a ranking on a set of items from noisy pairwise comparisons given item features. We address the fact that pairwise comparison data often reflects irrational choice, e.g. intransitivity. Our key observation is that two items compared in isolation from other items may be compared based on only a salient subset of features. Formalizing this framework, we propose the salient feature preference model and prove a finite sample complexity result for learning the parameters of our model and the underlying ranking with maximum likelihood estimation. We also provide empirical results that support our theoretical bounds and illustrate how our model explains systematic intransitivity. Finally we demonstrate strong performance of maximum likelihood estimation of our model on both synthetic data and two real data sets: the UT Zappos50K data set and comparison data about the compactness of legislative districts in the US.
APA
Bower, A. & Balzano, L.. (2020). Preference Modeling with Context-Dependent Salient Features. Proceedings of the 37th International Conference on Machine Learning, in Proceedings of Machine Learning Research 119:1067-1077 Available from https://proceedings.mlr.press/v119/bower20a.html.

Related Material