Does Preference Always Help? A Holistic Study on Preference-Based Evolutionary Multi-Objective Optimisation Using Reference Points

TL;DR

This study investigates whether incorporating user preferences in evolutionary multi-objective optimization consistently improves solution quality, revealing that effectiveness depends on proper preference elicitation and utilization.

Contribution

It provides a comprehensive overview of preference-based EMO and experimentally examines its effectiveness, highlighting the importance of proper preference elicitation and interactive methods.

Findings

Preference incorporation does not always improve solution approximation.

Invalid or poorly utilized preferences can hinder optimization.

Interactive preference elicitation can mitigate some issues.

Abstract

The ultimate goal of multi-objective optimisation is to help a decision maker (DM) identify solution(s) of interest (SOI) achieving satisfactory trade-offs among multiple conflicting criteria. This can be realised by leveraging DM's preference information in evolutionary multi-objective optimisation (EMO). No consensus has been reached on the effectiveness brought by incorporating preference in EMO (either a priori or interactively) versus a posteriori decision making after a complete run of an EMO algorithm. Bearing this consideration in mind, this paper i) provides a pragmatic overview of the existing developments of preference-based EMO; and ii) conducts a series of experiments to investigate the effectiveness brought by preference incorporation in EMO for approximating various SOI. In particular, the DM's preference information is elicited as a reference point, which represents her/his aspirations for different objectives. Experimental results demonstrate that preference incorporation in EMO does not always lead to a desirable approximation of SOI if the DM's preference information is not well utilised, nor does the DM elicit invalid preference information, which is not uncommon when encountering a black-box system. To a certain extent, this issue can be remedied through an interactive preference elicitation. Last but not the least, we find that a preference-based EMO algorithm is able to be generalised to approximate the whole PF given an appropriate setup of preference information.