Teams link https://teams.microsoft.com/meet/3673903989658?p=N6qTYL3GLXrbBWLnNm
Add to Google Calendar
Abstract
Semi-supervised learning (SSL) is a powerful approach for building predictive models that leverage both labeled and unlabeled data. While SSL has been extensively studied for relatively simple tasks such as classification and regression, its application to more complex prediction problems remains less explored. In many real-world scenarios, data exhibit inherent structure—such as in network-structured or spatial data—that SSL techniques can and should exploit. Similarly, in the target space, tasks such as multi-target regression, multi-label classification, and hierarchical multi-label classification introduce dependencies among output variables that SSL techniques should appropriately model and exploit.
In this seminar, we will introduce the fundamental concepts of SSL, discuss the challenges that arise when dealing with complex data, and explore several practical applications where the inherent structure of the data can be leveraged to improve learning performance.
Short Bio
Michelangelo Ceci, Ph.D., is a Full Professor of Computer Science at the University of Bari “Aldo Moro,” Italy. He has published more than 150 papers in leading journals and international conferences in the fields of Machine Learning and Data Mining. His research interests include advanced data analytics, complex structured data, and semi-supervised learning. He serves as unit coordinator for several European and national research projects, contributing to the design and coordination of large-scale initiatives in data science and artificial intelligence. Professor Ceci has been a program committee member for major conferences such as IEEE ICDM, SDM, IJCAI, and AAAI, and has served as Program Co-Chair for ECML-PKDD 2017 and 2026, SEBD 2007, Discovery Science 2016, ISMIS 2018, and ISMIS 2022. He was also the General Chair of ECML-PKDD 2017. In addition, he serves as an Associate Editor for the journals Machine Learning and Data Mining and Knowledge Discovery.
More information at News U. Porto