Finding optimal containment policies to balance GDP and mortality in a SEIARDS-V model using reinforcement learning.

Perin, Francesca (2022) Finding optimal containment policies to balance GDP and mortality in a SEIARDS-V model using reinforcement learning. Master's Thesis / Essay, Artificial Intelligence.

Preview	Text mAI_2022_PerinF.pdf Download (10MB) | Preview
	Text toestemming.pdf Restricted to Registered users only Download (145kB)

Abstract

The recent pandemic of COVID-19 renewed interest in epidemic models, with the scope of prediction and finding strategies to reduce mortality. Many models focus on single countries, since containment policies are implemented at national level. However, country demographic and other factors may affect the model greatly. In this thesis, we define a SEAIRDS-V epidemiological model and combine it with a Gross Domestic Product (GDP) economic model. Both models are dependent on country demographics data, consisting on contact rates between different age groups, that simulate 26 countries in the European Union and United Kingdom. Furthermore, we include population migration between countries to simulate travelling. Migration is modelled int two different ways. One of them being a fixed percentage of the population for all countries, and one based on aviation data from the Eurostat database. Defined this environment, we use Reinforcement Learning (Temporal Difference and Actor-Critic model) to determine if an optimal policy to contain both mortality and loss of GDP is detected across countries.

Item Type:	Thesis (Master's Thesis / Essay)
Supervisor name:	Grossi, D. and Weerd, H.A. de
Degree programme:	Artificial Intelligence
Thesis type:	Master's Thesis / Essay
Language:	English
Date Deposited:	02 Feb 2023 08:32
Last Modified:	02 Feb 2023 08:32
URI:	https://fse.studenttheses.ub.rug.nl/id/eprint/29195

Actions (login required)

View Item