Modeling Social Learning Using Dyna-Q and Ant Colony Optimization
Renato Zimmermann
Pareto Undergraduate Journal of New Economists
Vol. 1 No. 6, Issue 2023
pp. 86 - pp. 101
Abstract
This paper introduces a novel way of modeling social learning in macroeconomics using techniques from model-based reinforcement learning and ant colony optimization. The work extends previous works in bounded rationality and social learning by providing tools to complement previously-distinct models in adaptive learning. We test these new techniques using simulations of job search and consumption. Results demonstrate that models fit using the proposed techniques can learn core economic behaviours given no information about the environment, but do not fully fit reward functions in line with rational expectations theory.
Renato Zimmermann is a master's student in the University of Toronto, having graduated from UTM in 2022 with a double degree in Economics and Computer Science. His work focuses on using cutting-edge computer science, both in the hardware and software level, to improve economic models. He is currently researching techniques to improve the speed of algorithms used in Macroeconomic models through the use of GPU acceleration. Renato also actively writes and contributes to open source software and hopes to continue doing so wherever he takes his career. His paper “Modeling Social Learning using Dyna-Q and Ant Colony Optimization” was written and submitted as a final year thesis to complete the ECO420 Research in Applied Economics course. His paper was later accepted and presented at the prestigious Carroll Round undergraduate dissertation competition at Georgetown University.
About the Author
Get in Touch
Renato Zimmerman - renato.zimmermann@mail.utoronto.ca