Advancing Remote Sensing with Machine Learning

Laura Rosa - Wageningen University & Research

Short-bio: Laura E. Cué la Rosa is a researcher at the Laboratory of Geo-information Science and Remote Sensing at Wageningen University & Research. Her work focuses on predicting near-future deforestation using satellite data and artificial intelligence. Dr. Cué la Rosa’s expertise includes remote sensing, tropical forests, land use monitoring, time series analysis, and deep learning. Her research aims to enhance environmental monitoring and conservation efforts through advanced technological approaches. 

A Tutorial on Grammars for Visualization and Visual Analytics

Nivan Ferreira - Federal University of Pernambuco

Abstract:The increase in data availability and the popularization of data science brought a continuously growing need to use visualizations to explore complex data. Creating new visualizations, however, is a difficult task that often requires extensive programming expertise. High-level grammars for authoring visualizations provide a systematic and flexible way to specify visualizations in a declarative manner, lowering the entry barrier for data analysis. Such grammars effectively assist in the exploration of the visualization design space and of complex datasets. The goal of this tutorial is to then provide a summary of recently proposed grammars, so that a broad audience (e.g., students, researchers, professionals) can better understand how they are designed and used in practice for visualization and visual analytics tasks.

Short-bio: Nivan is an Assistant Professor at the Center for Informatics (CIn) at the Universidade Federal de Pernambuco (UFPE) in Brazil. His research interests include information visualization, visual analytics, machine learning, and applied mathematics, with a particular emphasis on developing interactive systems and techniques for analyzing large-scale spatiotemporal datasets.

Graph Neural Network and Sheaves

Ana Tenório - Getulio Vargas Foundation

Abstract: Graph Neural Networks (GNNs) are widely used in machine learning for graph-structure date, but despite their success, they can be ineffective unless the input graphs have “homophily” characteristics (i.e., we have graphs in which two linked nodes are likely to share similar attributes). In this mini-course we will present a generalization of GNNs that solve this problem through sheaf theory. Although sheaves are an important object in Category Theory, we will present a simplified version and no background in Category Theory will be required. We will start with a brief discussion about GNNs and then focus on shaves and their use in machine learning.

Short-bio: Ana Tenório is a postdoctoral researcher at FGV-EMAp in Rio de Janeiro. She holds a bachelor's, master's, and doctoral degree in Mathematics from the Institute of Mathematics and Statistics (IME) at the University of São Paulo (USP). Currently, her research interest focuses on applications of sheaf theory in machine learning.

An introduction to network science

Suzana Santos - Federal University of ABC

Abstract: Several real world systems can be represented as networks (sets of components that interact with each other), such as biological networks, telecommunication networks, computer networks, and social networks. Many of these systems can not be fully understood in terms of individual components separately. Therefore, to understand the overall system behavior, it is fundamental to study patterns of interaction between components. When these interaction patterns exhibit non-trivial characteristics, the network is considered complex. Such networks are the object of study of network science. This mini-course will introduce this field, covering topics such as graph theory, network models, network metrics, real-world applications and tools for network analysis.

Short-bio: Suzana is an Assistant professor at the Center for Mathematics, Computing and Cognition at the Federal University of ABC, in Santo André - SP. She holds a bachelor's, master's, and doctoral degree in Computer Science from the Institute of Mathematics and Statistics (IME) at the University of São Paulo (USP). She did a research intership at the Laboratory of Probability, Statistics and Modeling at the Sorbonne University, in France, and was a postdoctoral fellow at PaccanaroLab at the School of Applied Mathematics (Getulio Vargas Foundation), in Rio de Janeiro. Her research focuses on developing statistical methods to analyze complex networks, as well as network medicine and machine learning methods for problems in medicine, biology, and pharmacology.