Jožef Stefan Institute, Slovenia
Biography: Sašo Džeroski is a scientific councillor at the Jozef Stefan Institute, Ljubljana, Slovenia. He is also a full professor at the Jozef Stefan International Postgraduate School. His research group investigates machine learning and data mining (including structured output prediction and automated modeling of dynamic systems) and their applications (in environmental sciences, incl. ecology/ecological modelling, and life sciences, incl. systems biology/systems medicine).
He has participated in many international research projects and coordinated three of them in the past. Most recently, he lead the FET Open XTrack project MAESTRA (Learning from Massive, Incompletely annotated, and Structured Data). He is currently one of the principal investigators in the FET Flagship Human Brain Project. He has been scientific and/or organizational chair of numerous international conferences, including ECML PKDD 2017, DS-2014, MLSB-2009/10, ECEM and EAML-2004, ICML-1999 and ILP-1997/99: ICML and ECML PKDD are two of the most prominent scientific events in the area of machine learning and data science.
Multi-Target Prediction with Trees and Tree Ensembles
Increasingly often, we need to learn predictive models from big or complex data, which may comprise many examples and many input/output dimensions. When more than one target variable has to be predicted, we talk about multi-target prediction. Predictive modeling problems may also be complex in other ways, e.g., they may involve incompletely/partially labelled data, as in semi-supervised learning, or data placed in a network context.
The talk will first give an introduction to the different tasks of multi-target prediction, such as multi-target classification and regression, hierarchical versions thereof, and versions of the tasks that involve additional complexity (such as semi-supervised multi-target regression and network-based hierarchical multi-label classification). It will continue to present methods for solving such tasks, in particular predictive clustering trees and ensembles thereof. Finally, it will present example applications of multi-target prediction in the life sciences, focusing on predictive modeling in virtual compound screening for drug repurposing.