Toolboxes & Code Repositories

gym-covid

A gym environment for a compartment model modelling the first COVID wave in Belgium.

Description:

A Python implementation of the COVID model by Abrams et al. (2021). This is a compartment model used to analyse the first wave of COVID that occured in Belgium starting from 01 March 2020. It explicitly models hospitalizations and ICU-hospitalizations, as well as contact-reduction for symptomatic people (as they have to stay at home). We implement both the stochastic (binomial) model as well as the ODE model described in the paper as Markov Decision Processes (MDPs) that follow the gym API.

More Information:

  • Related research challenge: Situated AI

  • Responsible Research Lead: Pieter Libin

  • Authors: Mathieu Reymond, Conor F. Hayes, Pieter Libin

  • Programming Language: Python

  • Link to the website: 

  • Link to the repository: https://github.com/plibin/gym-covid

  • Categories: Multi-agent systems, Reinforcement learning

Publications:

  • Reymond, M., Hayes, C. F., Willem, L., Rădulescu, R., Abrams, S., Roijers, D., Howley, E., Mannion, P., Hens, N., Nowé, A., Libin, P. (2024). Exploring the pareto front of multi-objective covid-19 mitigation policies using reinforcement learning. Expert Systems with Applications, 249, 123686. https://doi.org/10.1016/j.eswa.2024.123686