Real-time Multiattribute Bayesian Preference Elicitation with Pairwise Comparison Queries

Shengbo Guo, Scott Sanner
Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, PMLR 9:289-296, 2010.

Abstract

Preference elicitation (PE) is an important component of interactive decision support systems that aim to make optimal recommendations to users by actively querying their preferences. In this paper, we outline five principles important for PE in real-world problems: (1) real-time, (2) multiattribute, (3) low cognitive load, (4) robust to noise, and (5) scalable. In light of these requirements, we introduce an approximate PE framework based on TrueSkill for performing efficient closed-form Bayesian updates and query selection for a multiattribute utility belief state — a novel PE approach that naturally facilitates the efficient evaluation of value of information (VOI) heuristics for use in query selection strategies. Our best VOI query strategy satisfies all five principles (in contrast to related work) and performs on par with the most accurate (and often computationally intensive) algorithms on experiments with synthetic and real-world datasets.

Cite this Paper

BibTeX
@InProceedings{pmlr-v9-guo10b, title = {Real-time Multiattribute Bayesian Preference Elicitation with Pairwise Comparison Queries}, author = {Guo, Shengbo and Sanner, Scott}, booktitle = {Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics}, pages = {289--296}, year = {2010}, editor = {Teh, Yee Whye and Titterington, Mike}, volume = {9}, series = {Proceedings of Machine Learning Research}, address = {Chia Laguna Resort, Sardinia, Italy}, month = {13--15 May}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v9/guo10b/guo10b.pdf}, url = {https://proceedings.mlr.press/v9/guo10b.html}, abstract = {Preference elicitation (PE) is an important component of interactive decision support systems that aim to make optimal recommendations to users by actively querying their preferences. In this paper, we outline five principles important for PE in real-world problems: (1) real-time, (2) multiattribute, (3) low cognitive load, (4) robust to noise, and (5) scalable. In light of these requirements, we introduce an approximate PE framework based on TrueSkill for performing efficient closed-form Bayesian updates and query selection for a multiattribute utility belief state — a novel PE approach that naturally facilitates the efficient evaluation of value of information (VOI) heuristics for use in query selection strategies. Our best VOI query strategy satisfies all five principles (in contrast to related work) and performs on par with the most accurate (and often computationally intensive) algorithms on experiments with synthetic and real-world datasets.} }
Endnote
%0 Conference Paper %T Real-time Multiattribute Bayesian Preference Elicitation with Pairwise Comparison Queries %A Shengbo Guo %A Scott Sanner %B Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2010 %E Yee Whye Teh %E Mike Titterington %F pmlr-v9-guo10b %I PMLR %P 289--296 %U https://proceedings.mlr.press/v9/guo10b.html %V 9 %X Preference elicitation (PE) is an important component of interactive decision support systems that aim to make optimal recommendations to users by actively querying their preferences. In this paper, we outline five principles important for PE in real-world problems: (1) real-time, (2) multiattribute, (3) low cognitive load, (4) robust to noise, and (5) scalable. In light of these requirements, we introduce an approximate PE framework based on TrueSkill for performing efficient closed-form Bayesian updates and query selection for a multiattribute utility belief state — a novel PE approach that naturally facilitates the efficient evaluation of value of information (VOI) heuristics for use in query selection strategies. Our best VOI query strategy satisfies all five principles (in contrast to related work) and performs on par with the most accurate (and often computationally intensive) algorithms on experiments with synthetic and real-world datasets.
RIS
TY - CPAPER TI - Real-time Multiattribute Bayesian Preference Elicitation with Pairwise Comparison Queries AU - Shengbo Guo AU - Scott Sanner BT - Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics DA - 2010/03/31 ED - Yee Whye Teh ED - Mike Titterington ID - pmlr-v9-guo10b PB - PMLR DP - Proceedings of Machine Learning Research VL - 9 SP - 289 EP - 296 L1 - http://proceedings.mlr.press/v9/guo10b/guo10b.pdf UR - https://proceedings.mlr.press/v9/guo10b.html AB - Preference elicitation (PE) is an important component of interactive decision support systems that aim to make optimal recommendations to users by actively querying their preferences. In this paper, we outline five principles important for PE in real-world problems: (1) real-time, (2) multiattribute, (3) low cognitive load, (4) robust to noise, and (5) scalable. In light of these requirements, we introduce an approximate PE framework based on TrueSkill for performing efficient closed-form Bayesian updates and query selection for a multiattribute utility belief state — a novel PE approach that naturally facilitates the efficient evaluation of value of information (VOI) heuristics for use in query selection strategies. Our best VOI query strategy satisfies all five principles (in contrast to related work) and performs on par with the most accurate (and often computationally intensive) algorithms on experiments with synthetic and real-world datasets. ER -
APA
Guo, S. & Sanner, S.. (2010). Real-time Multiattribute Bayesian Preference Elicitation with Pairwise Comparison Queries. Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 9:289-296 Available from https://proceedings.mlr.press/v9/guo10b.html.

Related Material