1007. Eliciting additive preference models from heterogeneous preferences sources
Invited abstract in session MB-8: Preference Learning 1, stream Multiple Criteria Decision Aiding.
Monday, 10:30-12:00Room: Clarendon SR 2.08
Authors (first author is the speaker)
| 1. | Vincent Mousseau
|
| LGI, CentraleSupélec |
Abstract
Additive preference representation is standard in Multiple Criteria Decision Analysis, and learning such a preference model dates back from the UTA method [Jacquet-Lagrèze, Siskos, 1982]. In this seminal work, an additive piece-wise linear model is inferred from a learning set composed of pairwise comparisons. In this setting, the learning set is provided by a single Decision-Maker (DM), and an additive model is inferred to match the learning set. We extend this framework to the case where (i) multiple DMs with heterogeneous preferences provide part of the learning set, and (ii) the learning set is provided as a whole, without knowing which DM expressed each pairwise comparison. Hence, the problem amounts to inferring a preference model for each DM and simultaneously “discovering” the segmentation of the learning set. In this paper, we show that this problem is computationally difficult. We propose a mathematical programming based resolution approach to solve this Preference Learning and Segmentation problem (PLS), and also propose a heuristic to deal with large datasets. We study the performance of both algorithms through experiments using synthetic and real data. Moreover, we study the identifiability of the PLS problem, and propose, when two DMs are involved, a systematic elicitation procedure that enables to identify the two additive piecewise linear preference models.
Keywords
- Artificial Intelligence
- Decision Analysis
- Multi-Objective Decision Making
Status: accepted
Back to the list of papers