Develop a principal–agent version of the OC market model with a regulator as principal

Develop a principal–agent game formulation of the finite-agent greenhouse gas offset credit market model in which the regulator acts as the principal with its own objectives and regulated firms act as agents who adapt their strategies accordingly, and investigate the resulting equilibrium using reinforcement learning methods.

Background

The current work studies a finite-agent Nash equilibrium among firms in the OC market using Nash-DQN. The authors highlight extending the framework to a principal–agent setting where the regulator’s objectives influence firm strategies.

Prior RL-based contract theory work is cited as evidence of feasibility, suggesting that integrating regulatory objectives could yield a richer game-theoretic structure and more realistic policy modeling.