Hypernetwork-based approach for optimal composition design in partially controlled multi-agent systems

TL;DR

This paper introduces a hypernetwork-based framework for optimizing agent composition and policies in partially controlled multi-agent systems, significantly improving efficiency and performance in real-world scenarios like taxi dispatching.

Contribution

The study presents a novel hypernetwork approach that jointly optimizes composition and policies, reducing computational costs compared to traditional separate training methods.

Findings

Outperforms existing methods in approximating equilibrium policies

Achieves higher order response rate and served demand

Demonstrates practical utility in real-world data

Abstract

Partially Controlled Multi-Agent Systems (PCMAS) are comprised of controllable agents, managed by a system designer, and uncontrollable agents, operating autonomously. This study addresses an optimal composition design problem in PCMAS, which involves the system designer's problem, determining the optimal number and policies of controllable agents, and the uncontrollable agents' problem, identifying their best-response policies. Solving this bi-level optimization problem is computationally intensive, as it requires repeatedly solving multi-agent reinforcement learning problems under various compositions for both types of agents. To address these challenges, we propose a novel hypernetwork-based framework that jointly optimizes the system's composition and agent policies. Unlike traditional methods that train separate policy networks for each composition, the proposed framework generates policies for both controllable and uncontrollable agents through a unified hypernetwork. This approach enables efficient information sharing across similar configurations, thereby reducing computational overhead. Additional improvements are achieved by incorporating reward parameter optimization and mean action networks. Using real-world New York City taxi data, we demonstrate that our framework outperforms existing methods in approximating equilibrium policies. Our experimental results show significant improvements in key performance metrics, such as order response rate and served demand, highlighting the practical utility of controlling agents and their potential to enhance decision-making in PCMAS.