From Optimal Policies to Individual Differences: Rethinking Reinforcement Learning for Biology
Neural and Evolutionary Computing
Summary
The authors explain that reinforcement learning (RL) is often used to model how living things learn and make decisions. However, current RL models usually show average behavior or one ideal example, missing the natural differences between individuals seen in real life. They explore different RL methods from various fields to find ways to create more realistic, varied behaviors. Their goal is to better match computer simulations with the diverse actions observed in biology.
Authors
Patrick Govoni, Palina Bartashevich, Clémence Bergerot, Valerii Chirkov, Valentin Lecheval, Pawel Romanczuk
Abstract
Reinforcement learning (RL) is primarily known as a computational method for optimizing control tasks, but it is increasingly used to explain biological behavior. While RL successfully captures key aspects of biology, a major gap remains: between-agent behavioral variability. Consistent individual differences naturally permeate biological populations, yet RL models typically present only the single best individual or the population average. Addressing this gap requires moving beyond current practices to generate behavioral diversity using biologically plausible mechanisms. Here, we examine approaches from various subfields of RL and outline potential paths forward to close the gap between biology and simulation.