I was educated at the University of Patras, Greece, in Electrical and Computer Engineering (BEng/MEng, PhD). I hold a Post Graduate Certificate in Teaching and Learning in Higher Education from Brunel University, UK. I have been teaching and researching in the area of data-driven modelling and machine learning since 1993. My research activities fall under the umbrella of intelligent technologies and involve key data and information processing paradigms, such as fuzzy logic, neural learning, and global search. The aim of this work is the development of intelligent machines that exhibit different levels of learning, seamlessly combine explicit knowledge representation with significant learning from data capabilities, and can handle uncertainty which is inherent in complex environments. The above constitute a conceptual framework within which more specific topics are pursued, and there are several well defined, coherent research streams:
Neural networks and learning from data systems - I have been researching in this area since 1993. My work has focused on supervised learning, dimensionality reduction, transfer learning and pattern classification. I have developed new learning algorithms that provide improved learning, studied the convergence behaviour of existing and new learning algorithms, and applied advanced training schemes in applications in medical imaging, microarray image analysis and bioinformatics. My work started with deterministic approaches to learning based on optimisation theory but gradually shifted towards more stochastic and hybrid techniques, such as the new approaches inspired by the theory of nonextensive statistics and the recent advances in behavioural genetics. The last few of years I work, together with my PhD students, for developing effective learning methods for dynamic networks and other high-order connectionist architectures, ensembles learning and learning in changing environments.
Methods for global searching, adaptation and evolution - I have been researching in this since completing my PhD. My work in this area has concentrated on new approaches to learning in adaptive systems through optimisation, and includes data-driven algorithms that combine adaptation and evolution to solve problems in complex environments; in particular nature-inspired methods, like genetic and evolutionary computing, swarms, and differential evolution. Nature-inspired methods can help addressing some very important research challenges today, and enable us to design machines with a higher level of flexibility and autonomy that will be able to learn from data, develop their understanding of the environment and, ultimately, their intelligence.
Adaptive and personalised systems - my work in this area started in 2000 and has been expanded with my involvement in the London Knowledge Lab when I joined Birkbeck in 2004. The main directions here are user-behaviour modelling and data-driven adaptation, personalisation and control. Applications are in educational technology and web-based systems. My work in this area has led so far to a series of research grants from UK funding agencies, including the EPSRC, the ESRC, the AHRC and the JISC.
My research work has received awards from the ACM (in 2009, for work in Evolutionary Computation), the IEEE (in 2000 and in 2008, for work in Neural Networks Learning and in Feature Selection respectively), the EUNITE (in 2001 and in 2004, for work on Neural Networks for Medical Imaging and Bioinformatics respectively) and the IADIS (in 2006, for work in personalised systems).
Before joining academia I held R&D positions in industry, between 1990 and 1993 with AMBER S.A (a subsidiary of the HERACLES General Cement Company, Athens, Greece - member of the Lafarge Group) and between 1997 and 1998 with SYNDESIS Ltd, where I worked in several national and international projects on the development of embedded systems employing soft computing and computational intelligence methodologies. In 1998, I took up the post of research fellow in the Department of Informatics and Telecommunications at the University of Athens where I participated on EU-funded projects, and national projects in the area of medical imaging and educational technology. This was followed in 1999 by a postdoctoral fellowship (funded by the Greek State Scholarships Foundations) in the Department of Mathematics at the University of Patras to research new approaches to learning in adaptive systems through optimisation. In 2000, I joined the Department of Information Systems and Computing at Brunel University, where I worked first as a Lecturer and later on as a Senior Lecturer. In 2004, I moved to Birkbeck College, University of London, where I am currently Professor of Computer Science in the Department of Computer Science and Information Systems.
I have (co-)edited three books, entitled Adaptable and Adaptive Hypermedia Systems (2005), Advances in Web-based Education: Personalized Learning Environments (2006), E-Infrastructures and Technologies for Lifelong Learning (2011), and Investigations into Living Systems, Artificial Life, and Real-World Solutions (2013). I am a member of the EPSRC College, the IEEE, the User Modeling Inc, the International Artificial Intelligence in Education Society and a honorary member of the Hellenic Artificial Intelligence Society, and a Charted Engineer, member of the Technical Chamber of Greece since 1990.