EpidemiOptim: A Toolbox for the Optimization of Control Policies in Epidemiological Models

• Published in: IDEAS Working Paper Series from RePEc Vol. 71; pp. 479 - 519
• Main Authors: Colas, Cédric, Hejblum, Boris, Rouillon, Sebastien, Thiébaut, Rodolphe, Oudeyer, Pierre-Yves, Moulin-Frier, Clément, Prague, Mélanie
• Format: Journal Article Paper
• Language: English
• Published: San Francisco AI Access Foundation 2021; Federal Reserve Bank of St. Louis; Association for the Advancement of Artificial Intelligence
• Subjects: Algorithms; Artificial Intelligence; Artificial neural networks; Computer Science; Coronaviruses; Cost function; COVID-19; Decision making; Deep learning; Disease control; Economic models; Epidemics; Epidemiology; Evolutionary algorithms; Fatalities; Life Sciences; Machine Learning; Modelling; Optimization; Policies; Quantitative Methods; Decision Making; Visualization; Cost Functions; Epidemiological Models; Control Strategies; Optimization Algorithms; Reinforcement Learning; Learning Systems; Deep Neural Networks; Epidemiology; Optimization; Learning Algorithms; Deep Learning; Non-Pharmaceutical Interventions; Optimization Problems; Interactive Visualizations; Standard Interface; Evolutionary Algorithms; State-Of-The-Art Machine Learning Methods
• ISSN: 1076-9757; 1943-5037; 1076-9757; 1943-5037
• DOI: 10.1613/jair.1.12588

Modeling the dynamics of epidemics helps to propose control strategies based on pharmaceuticaland non-pharmaceutical interventions (contact limitation, lockdown, vaccination,etc). Hand-designing such strategies is not trivial because of the number of possibleinterventions and the difficulty to predict long-term effects. This task can be cast as an optimization problem where state-of-the-art machine learning methods such as deep reinforcement learning might bring significant value. However, the specificity of each domain|epidemic modeling or solving optimization problems|requires strong collaborationsbetween researchers from different fields of expertise. This is why we introduce EpidemiOptim, a Python toolbox that facilitates collaborations between researchers inepidemiology and optimization. EpidemiOptim turns epidemiological models and cost functions into optimization problems via a standard interface commonly used by optimization practitioners (OpenAI Gym). Reinforcement learning algorithms based on QLearning with deep neural networks (DQN) and evolutionary algorithms (NSGA-II) are already implemented. We illustrate the use of EpidemiOptim to find optimal policies fordynamical on-o  lockdown control under the optimization of the death toll and economic recess using a Susceptible-Exposed-Infectious-Removed (SEIR) model for COVID-19. Using EpidemiOptim and its interactive visualization platform in Jupyter notebooks, epidemiologists, optimization practitioners and others (e.g. economists) can easily compare epidemiological models, costs functions and optimization algorithms to address important choicesto be made by health decision-makers. Trained models can be explored by experts and non-experts via a web interface. This article is part of the special track on AI and COVID-19.