SingularityNET: Agent-Based Simulation of COVID-19 Health and Economical Effects – Dr. Petronio Candido

➡️ COVID-19 Simulation Summit Playlist:
👀 About the speaker
Petrônio Cândido L. Silva is a Brazilian professor, developer, and researcher. He has a B.Sc degree in Information Systems(2005), an M.Sc degree in Informatics (2010), and a Ph.D. in Electrical Engineering(2019). Currently, he works in the Federal Institute of Minas Gerais (Brazil) and is a member of MINDS – Machine Intelligence and Data Science Laboratory (at UFMG/Brazil).


Speaker’s Abstract:
“Agent-Based Simulation (ABS) is a good choice to simulate dynamic complex systems due to its simplicity of implementation and accurate results when compared with real data. On the other hand, SIR and SEIR models are commonly used to estimate the time evolution of the COVID-19 disease in epidemiological terms. However, as we add new variables in the simulations, their interactions become more complex and hard to describe in analytical terms. In this presentation, we show how ABS can be employed in the simulation of complex and dynamic scenarios with many interacting epidemiological, social, and economic variables.”

On April 30th, 2020, the DAIA Foundation has organized an online COVID-19 Simulation Summit. The summit was focused on the use of agent based simulation models for more effective simulation of COVID-19 spread and evaluation of COVID-19 policies. Consisting of live video talks, Q&A sessions, and panel discussions,
the event gathered together scientists with insight and experience in simulation modeling of complex systems (especially but not only agent based modeling) and complex systems dynamics; along with scientists and physicians with specific insight into COVID-19 and related epidemiological issues.

The aim of the COVID-19 Simulation Summit is to gain understanding of
What tools and approaches should be used to provide the most accurate possible models of COVID-19 and its spread and implications, both in the current phase and in future phases of disease propagation?
What assumptions should be made or questioned within COVID-19 simulation models?
What assumptions have been left unquestioned in current models and need to be more thoroughly explored?
What data sources might best be leveraged in constructing models related to COVID-19?
What data should we be gathering that we aren’t currently, and how may this be done in a way that respects personal privacy and data sovereignty?
What real-world test cases might best be used for initial exploration of more accurate simulation models
Agent-Based Simulation
The epidemiological models of COVID-19 spreading dynamics that are driving current policy decisions are generally well thought out and carefully implemented, however they also tend to be highly simplistic relative to the complexity of the actual situation.

Among many other factors, they don’t generally factor in the impacts of various interventions and control policies, nor do they account for the different behavior patterns of different classes of people.

The agent based simulation paradigm allows a finer-grained sort of modeling, in which a region (or the world as a whole) is modeled as a specific geometry occupied by interacting autonomous agents with a diversity of specific behavior patterns. An in-depth agent based simulation of COVID-19 spreading would allow better-grounded policy choices to be made regarding how to manage, control and cope with the pandemic.

An agent based simulation, like any other model, depends on the underlying assumptions used to structure it. However, the agent based modeling paradigm provides a more flexible approach to evaluating the consequences of various assumptions, and thus exploring their validity.


SingularityNET is a decentralized marketplace for artificial intelligence. We aim to create the world’s global brain with a full-stack AI solution powered by a decentralized protocol.

We gathered the leading minds in machine learning and blockchain to democratize access to AI technology. Now anyone can take advantage of a global network of AI algorithms, services, and agents.


YouTube Source for this SingularityNET AI Video

AI video(s) you might be interested in …