João Gama is Associate Professor of the Faculty of Economy, University of Porto. He is a researcher and vice-director of LIAAD, a group belonging to INESC TEC. He got the PhD degree from the University of Porto, in 2000. He has worked in several National and European projects on Incremental and Adaptive learning systems, Ubiquitous Knowledge Discovery, Learning from Massive, and Structured Data, etc.
He served as Co-Program chair of ECML'2005, DS'2009, ADMA'2009, IDA' 2011, and ECML/PKDD'2015. He served as track chair on Data Streams with ACM SAC from 2007 till 2016. He organized a series of Workshops on Knowledge Discovery from Data Streams with ECML/PKDD, and Knowledge Discovery from Sensor Data with ACM SIGKDD.
He is author of several books in Data Mining (in Portuguese) and authored a monograph on Knowledge Discovery from Data Streams. He authored more than 250 peer-reviewed papers in areas related to machine learning, data mining, and data streams.
He is a member of the editorial board of international journals ML, DMKD, TKDE, IDA, NGC, and KAIS. He supervised more than 12 PhD students and 50 Msc students.
Title: Real-Time Data Mining
Nowadays, there are applications in which the data are modelled best not as persistent tables, but rather as transient data streams. In this keynote, we discuss the limitations of current machine learning and data mining algorithms. We discuss the fundamental issues in learning in dynamic environments like learning decision models that evolve over time, learning and forgetting, concept drift and change detection. Data streams are characterized by huge amounts of data that introduce new constraints in the design of learning algorithms: limited computational resources in terms of memory, processing time and CPU power. In this talk, we present some illustrative algorithms designed to taking these constrains into account. We identify the main issues and current challenges that emerge in learning from data streams, and present open research lines for further developments.
Roberto Santana received the M.Sc. degree in Computer Science from the University of Havana, Cuba, in 1996. He received a Ph.D. in Mathematics from the University of Havana in 2005 and a Ph. D. in Computer Science from the University of the Basque Country, in Spain, in 2006.
He is a Tenured researcher at the Intelligent Systems Group (ISG), Department of Computer Science and Artificial Intelligence, University of the Basque Country. Roberto Santana's research interests comprise the use of probabilistic graphical models in evolutionary algorithms and the application of machine learning methods to problems from bioinformatics and neuroinformatics.
He has published over 25 papers in international journals, more than 60 papers in international conferences, and is the editor of one book devoted to the use of Markov networks in evolutionary computation. He has participated in several projects funded by the Basque Government, the Spanish Ministry of Science and Technology, and the European Union.
He is also a professor in the Master program "Computational Engineering and Intelligent Systems" of the University of the Basque Country and currently the co-supervisor of four Ph.D projects, four Master projects, and a postdoctoral student. Previously, Santana has been a member of the Computational Intelligence Group (CIG) in the Technical University of Madrid, and the Institute of Cybernetics, Mathematics and Physics (ICIMAF) of Havana.
He has been a visiting researcher in the Institute Autonomous Intelligent Systems (AIS) in Bonn, Germany; the Centre for Informatics and Systems of the University of Coimbra, Portugal; the University of Wellington, New Zealand; the Ivane Javakhishvili Tbilisi State University in Georgia; and the Jiao-Tong University of Shanghai, China.
He is currently an CNPq Special Visiting Researcher (Pesquisador Visitante Especial) in the Federal University of Paraná. Roberto Santana has been invited to plenary talks and tutorials in Colombia and the United Kingdom.
Title: Estimation of distribution algorithms: Competent evolutionary algorithms, knowledge extraction techniques, or an arena for testing problem structural hypotheses and machine learning methods?
Estimation of distribution algorithms (EDAs) were originally conceived as a more efficient class of evolutionary approaches based on building probabilistic models of the search space. Traditionally, machine learning methods are used to learn these models from the solutions evaluated along the evolution, and to sample new promising solutions from the models. EDAs have been successfully applied to a variety of practical problems in Bioinformatics, energy production, industrial scheduling, etc. This talk discusses recent developments, extensions, and relevant results on the theory and application of EDAs. In particular, results on the application of these algorithms to unveil previously unknown information about the structure of the addressed optimization problem are reviewed. Examples of EDAs applications to real-world optimization problems are discussed. The link between EDAs, other model-based evolutionary algorithms, and research areas such as knowledge-discovery, data mining, and transfer learning are covered in the presentation. Finally, some open and challenging problems on EDAs are briefly presented.