Policy-focused Agent-based Modeling using RL Behavioral Models

TL;DR

This paper explores the use of reinforcement learning models as behavioral agents in policy-oriented ABMs, demonstrating their effectiveness and superiority over traditional models in two complex simulation scenarios.

Contribution

It introduces RL-based behavioral models for ABMs, extending RL algorithms to multi-agent settings, and validates their effectiveness in policy-relevant simulations.

Findings

RL agents effectively maximize rewards in ABMs.

RL models outperform traditional adaptive behavioral models.

RL-based ABMs can handle multi-agent interactions and heterogeneity.

Abstract

Agent-based Models (ABMs) are valuable tools for policy analysis. ABMs help analysts explore the emergent consequences of policy interventions in multi-agent decision-making settings. But the validity of inferences drawn from ABM explorations depends on the quality of the ABM agents' behavioral models. Standard specifications of agent behavioral models rely either on heuristic decision-making rules or on regressions trained on past data. Both prior specification modes have limitations. This paper examines the value of reinforcement learning (RL) models as adaptive, high-performing, and behaviorally-valid models of agent decision-making in ABMs. We test the hypothesis that RL agents are effective as utility-maximizing agents in policy ABMs. We also address the problem of adapting RL algorithms to handle multi-agency in games by adapting and extending methods from recent literature. We evaluate the performance of such RL-based ABM agents via experiments on two policy-relevant ABMs: a minority game ABM, and an ABM of Influenza Transmission. We run some analytic experiments on our AI-equipped ABMs e.g. explorations of the effects of behavioral heterogeneity in a population and the emergence of synchronization in a population. The experiments show that RL behavioral models are effective at producing reward-seeking or reward-maximizing behaviors in ABM agents. Furthermore, RL behavioral models can learn to outperform the default adaptive behavioral models in the two ABMs examined.