Reverse emulating agent-based models for policy simulation

This is an idea proposed in 2023 as a Cambridge Computer Science Part III or MPhil project, and has been completed by Pedro Sousa. It was co-supervised with Sadiq Jaffer.

Governments increasingly rely on simulation tools to inform policy design. Agent-based models (ABMs) simulate complex systems to study the emergent phenomena of individual behaviours and interactions in agent populations. However, these ABMs force an iterative, time-consuming, unmethodical parameter tuning of key policy "levers" (or input parameters) to steer the model towards the envisioned outcomes. To unlock a more natural workflow, this project investigates reverse emulation, a novel approach that streamlines policy design using probabilistic machine learning to predict parameter values that yield the desired policy outcomes.

Background reading

• J. Dyer, P. Cannon, J. D. Farmer, and S. M. Schmon, "Black-box bayesian inference for agent-based models", Journal of Economic Dynamics and Control, vol. 161, p. 104827, 2024.
• E. Frias-Martinez, G. Williamson, and V. Fr ́ıas-Mart ́ınez, "An agent-based model of epidemic spread using human mobility and social network information," pp. 57–64, 10 2011.

Links

• Publication to follow as it is currently being written up. The project was awarded the "2024 Highly Commended M.Phil Project" commendation from the Computer Science department.

See Also

This project was a followup to one in the previous year by Sharan Agrawal on Scalable agent-based models for optimized policy design.