From Preference-Based to Multiobjective Sequential Decision-Making

TL;DR

This paper establishes a theoretical connection between preference-based and multiobjective sequential decision-making, enabling the transfer of solution methods and addressing problems with unknown reward values.

Contribution

It introduces a method to transform preference-based problems into multiobjective ones under certain conditions, expanding the scope of applicable decision-making techniques.

Findings

Provides a formal link between the two decision-making frameworks.

Justifies using multiobjective methods for preference-based problems with unknown rewards.

Enables new problem formulations in multiobjective decision-making.

Abstract

In this paper, we present a link between preference-based and multiobjective sequential decision-making. While transforming a multiobjective problem to a preference-based one is quite natural, the other direction is a bit less obvious. We present how this transformation (from preference-based to multiobjective) can be done under the classic condition that preferences over histories can be represented by additively decomposable utilities and that the decision criterion to evaluate policies in a state is based on expectation. This link yields a new source of multiobjective sequential decision-making problems (i.e., when reward values are unknown) and justifies the use of solving methods developed in one setting in the other one.