AI4GCC -- Track 3: Consumption and the Challenges of Multi-Agent RL

TL;DR

This paper discusses the AI4GCC competition's integration of machine learning with economic policy analysis, emphasizing the need for improved evaluation metrics and understanding agent learning dynamics in multi-agent reinforcement learning scenarios.

Contribution

It proposes adding consumption/utility metrics and investigating agent learning dynamics and game-theoretic properties to enhance the competition's effectiveness.

Findings

Suggested inclusion of consumption/utility index in evaluation

Highlighted importance of studying agent learning dynamics

Emphasized analyzing game-theoretic properties of outcomes

Abstract

The AI4GCC competition presents a bold step forward in the direction of integrating machine learning with traditional economic policy analysis. Below, we highlight two potential areas for improvement that could enhance the competition's ability to identify and evaluate proposed negotiation protocols. Firstly, we suggest the inclusion of an additional index that accounts for consumption/utility as part of the evaluation criteria. Secondly, we recommend further investigation into the learning dynamics of agents in the simulator and the game theoretic properties of outcomes from proposed negotiation protocols. We hope that these suggestions can be of use for future iterations of the competition/simulation.