Dr. Dejan Petelin

web: http://dejan.petelin.si/

Details of education

• 2009 Diploma degree at the Faculty of Computer and Information Science, University of Ljubljana, Slovenia
• from 2009 onwards postgraduate student at Jožef Stefan International Postgraduate School
• 2014 Ph. D. at Jožef Stefan Insternational Postgraduate School, Slovenia

Career to date

• from 2009 onwards employed as a junior researcher at Jožef Stefan Institute
• from 2014 postdoctoral associate at Jožef Stefan Institute

Research interests

My Ph.D. research is focused on Evolving Gaussian process (GP) models and their aplicability for various applications such as control system, prognostics, multi-objective optimization, etc. Evolving GP models are inspired by Evolving systems that have both their structure and their parameters self-adapting on-line.

Honours & Awards

• Second prize at the IEEE PHM 2012 Prognostic Challenge (details & results, paper)
• Industry collaboration award given by Jožef Stefan International Postgraduate School (2011)
• Jožef Stefan Institute grant (2004 - 2009)

Selected references:

1. PETELIN, Dejan, GRANCHAROVA, Alexandra, KOCIJAN, Juš. Evolving Gaussian process models for prediction of ozone concentration in the air. Simulation Modelling Practice and Theory, Elsevier, 2013, vol. 33, 68-80. link
2. BOŠKOSKI, Pavle, GAŠPERIN, Matej, PETELIN, Dejan. Bearing fault prognostics using Rényi entropy based features and Gaussian process models. Mechanical Systems and Signal Processing, Elsevier, 2013, Submitted.
3. MLAKAR, Miha, PETELIN, Dejan, TUŠAR, Tea, FILIPIČ, Bogdan. GP-DEMO: Differential evolution for multiobjective optimization based on Gaussian process models. European Journal of Operational Research, Elsevier, 2013, Submitted.
4. BOŠKOSKI, Pavle, GAŠPERIN, Matej, PETELIN, Dejan. Bearing fault prognostics based on signal complexity and Gaussian process models. Prognostics and Health Management (PHM), 2012 IEEE Conference on, 2012. link
5. PETELIN, Dejan, KOCIJAN, Juš. Control System with Evolving Gaussian Process Models. 2011 IEEE Workshop on Evolving and Adaptive Intelligent Systems, 2011. link
6. KOCIJAN, Juš, PETELIN, Dejan. Output-error model training for Gaussian process models. Adaptive and Natural Computing Algorithms, 10th International Conference, ICANNGA 2011, Proceedings, Part II, Springer, 2011. link
7. PETELIN, Dejan, FILIPIČ, Bogdan, KOCIJAN, Juš. Optimization of Gaussian process models with evolutionary algorithms. Adaptive and Natural Computing Algorithms, 10th International Conference, ICANNGA 2011, Proceedings, Part I, Springer, 2011. link
8. MUSIZZA, Bojan, PETELIN, Dejan, KOCIJAN, Juš. Accelerated learning of Gaussian process models. EUROSIM 2010 : proceedings of the 7th EUROSIM Congress on Modelling and Simulation, September 6-10, 2010, Prague, Czech Republic. Vol. 2, Prague: Czech Technical University in Prague, 2010, 7 pages.

Complete bibliography: available at COBISS