Decision Maker Preferences in Surrogate-based Multi-Objective Optimization: A Survey

Sasan Amini, Antonio Candelieri, Inneke Van Nieuwenhuyse

Confirmation: I have read and agree with the workshop's policy on behalf of myself and my co-authors.

Tracks: Main Track

Keywords: Multi-objective optimization, Surrogate-based optimization, Pareto-front, Preferences

TL;DR: This work discusses how the incorporation of the decision-maker preferences can contribute to the main goal of increasing data efficiency in surrogate-based algorithms.

Abstract: Multi-objective optimization problems are highly relevant in practice, and algorithms to solve these types of problems abound in the literature. This survey focuses explicitly on surrogate-based algorithms that use the decision-maker's preference information to guide the search toward the most preferred areas of the Pareto front. Considering such preferences not only facilitates the decision-making process for the user but also helps the analyst to save expensive computational budget. The way in which user preference information is handled in the algorithms differs across publications. We classify them according to the type and timing of the preference information. We provide an overview of the state-of-the-art, highlight the most important shortcomings in the literature, and present promising directions for further research.

Submission Number: 29