04/08/2021

Ulcigrai and Nonato are featured at the Colloquium.

On the third day of the 3Tuesdays Brazilian Mathematics Colloquium, prominent names in national and international mathematics were featured in the program. While Italian researcher Corinna Ulcigrai spoke about parabolic dynamical systems that exhibit a slow form of "butterfly effect," mathematician Luis Gustavo Nonato explained how data science can be used in the development of more efficient public policies.

Corinna Ulcigrai researches ergodic theory, dynamical systems, and Teichmueller dynamics, focusing on what she calls "slowly chaotic systems," aiming to discover mechanisms that make even simple systems chaotic and difficult to predict, and to develop mathematical tools that can help understand them. In her plenary lecture "Parabolic flows and slow chaos," she reviewed some of the recent advances in understanding typical chaotic characteristics and the main mechanisms that explain them, achieved through the exploration of a mixture of dynamical, analytical, and geometric techniques.

A professor at the University of Zurich, the Italian scientist received the European Mathematical Society Prize in 2012 and the Whitehead Prize in 2013. In 2020, she won the Michael Brin Prize in Dynamical Systems, the most important distinction in the field. During the event, held virtually this year, she interacted with participants who sent comments and thanked them for participating in the Colloquium: "It's a shame I can't be with you in person. But if the event weren't online, with my activities, I probably wouldn't be able to be here today. So, there are also advantages to this digital format."

Luis Gustavo Nonato, a researcher at the Institute of Mathematical and Computer Sciences at the University of São Paulo, spoke about a project he coordinates that seeks to identify crime patterns and create tools to predict future cases. In his presentation titled "Data Intelligence for Public Security," he explained how numerous factors, such as traffic and weather conditions, population density, and socioeconomic conditions, influence the number of criminal occurrences. The mathematician also detailed how the analysis of this information can help in the development of more efficient public policies.

“Data science is revolutionizing the way public policies are planned worldwide, and this is also beginning to happen in Brazil. We have been receiving calls from many security departments, government agencies, and even the judiciary, asking for help to optimize processes and improve the efficiency of public policies based on evidence,” Nonato highlighted. He also noted that the area still has a lot of demand, but there are still few specialized professionals.

The researcher also spoke about how he decided to study data science while a visiting professor at New York University, after a lecture on a project to improve health inspections in Chicago: “There were very few inspectors and a huge number of restaurants. The solution was to set up a machine learning system based on information from restaurant review apps to rank establishments according to negative comments in order to allocate inspectors where the need was greatest. In less than a year, the number of highly rated restaurants grew enormously. And this mechanism began to be used in various areas: to prevent fires, to prevent episodes of police violence, for example. This is a relatively simple solution, but one that can have a very large social impact.”

The 3Tuesdays Brazilian Mathematics Colloquium will be held virtually until next Friday (06) and includes activities such as plenary and outreach lectures, as well as courses. The videos of the presentations are available on the event's YouTube page .