Prof. Antônio de Pádua Braga
Antônio de Pádua Braga received a B.Sc degree in Electrical Engineering (1987) and a Master Degree in Computer Science (1991), both from Federal University Minas Gerais, Brazil. His Ph.D. on Electrical Engineering was obtained from the University of London, Imperial College, in 1995, in Recurrent Neural Networks. Since 1991, he has been with the Electronics Engineering Department at Federal University of Minas Gerais where he is now a Full Professor and head of the Computational Intelligence Laboratory. He is also an Associate Researcher of the Brazilian National Research Council (CNPq).
As a Professor and researcher he has co-authored many books, book-chapters, journal and conference papers. As a research leader he has received many grants from Brazilian government research agencies and from private companies. Among the research grants received two were in the frame of international research cooperation agreements. One of them was with University of Paris, France, supported by the agencies CAPES (Brazil) and COFECUB (France) and the other with Université catholic de Louvain, Belgium, supported by CNPq (Brazil) and FNRs (Belgium).
He is now an Associate Editor of the international journal IEEE Transactions on Neural Networks and Learning Systems (2014-2015 term). He is also in the editorial board of the International Journal of Computational Intelligence and Applications, published by Imperial College Press, of which he was co-editor in chief and founder. He has served in many program committees of various international conferences, among them: International Joint Conference on Neural Networks (IJCNN), European Symposium on Neural Networks (ESANN), International Conference on Neural Networks (ICANN) and European Control Conference (ECC).
Although our publications vary from fundamental topics to applied ones, our main research interest is on the basic issues related to Supervised and Semi-Supervised Learning, mainly with Artificial Neural Networks. Our group was able to formally describe a Multi-Objective Optimization approach to neural networks learning in the early 1990s. Since then, most of our research activities were aimed at this topic. Presently, our main interest in the area is on the development of new decision-making strategies for Pareto set model selection of neural networks. Despite of Multi-objective learning being our main research topic in the last years, we have also interest on other approaches of learning. In a more recently work we have been investigating how prior structural information embedded in affinity matrices can serve as a regularizer for neural networks and extreme learning machines. Other aspects of machine learning are within the scope of our interests, such as supervised and semi-supervised feature selection, active learning, kernel machines and large margin classifiers. We have also been working with applications of artificial neural networks in areas such as modeling of industrial processes, energy systems and bioinformatics.