Navigating Trade-offs: Policy Summarization for Multi-Objective Reinforcement Learning

TL;DR

This paper introduces a clustering method for multi-objective reinforcement learning solutions, helping decision makers understand trade-offs and policy behaviors more effectively than traditional clustering methods.

Contribution

The paper presents a novel clustering approach that considers both policy behavior and objective values to analyze MORL solution sets, improving interpretability.

Findings

Outperforms traditional k-medoids clustering in four environments

Reveals relationships between policy behaviors and objective space regions

Demonstrates practical application through a real-world case study

Abstract

Multi-objective reinforcement learning (MORL) is used to solve problems involving multiple objectives. An MORL agent must make decisions based on the diverse signals provided by distinct reward functions. Training an MORL agent yields a set of solutions (policies), each presenting distinct trade-offs among the objectives (expected returns). MORL enhances explainability by enabling fine-grained comparisons of policies in the solution set based on their trade-offs as opposed to having a single policy. However, the solution set is typically large and multi-dimensional, where each policy (e.g., a neural network) is represented by its objective values. We propose an approach for clustering the solution set generated by MORL. By considering both policy behavior and objective values, our clustering method can reveal the relationship between policy behaviors and regions in the objective space. This approach can enable decision makers (DMs) to identify overarching trends and insights in the solution set rather than examining each policy individually. We tested our method in four multi-objective environments and found it outperformed traditional k-medoids clustering. Additionally, we include a case study that demonstrates its real-world application.