Dr. Peter Sykacek Dept. of Biotechnology Boku University, Vienna Muthgasse 18 A-1190 Wien, AUSTRIA Phone: +43-1-36006-6836 Email: peter@sykacek.net Web: http://www.sykacek.net/ Curriculum Vitae Dr. Peter Sykacek Objective Addendum to Grant Proposal Experience 01/2006Boku University Vienna Group Leader Analytical Methods in Computational Biology Vienna, AT Group Leader in the Bioinformatics Group, Department of Biotechnology, Boku University, Vienna. I am currently establishing a research group with a focus on analytical methods development in computational biology. We are specialised in the application of highly structured stochastic systems to optimally integrated models for analysis of molecular biological data. 04/2004-12/2005 University of Cambridge Postdoctoral Research Fellow Cambridge, UK Senior statistician in Dr. Micklem’s Group in the Department of Genetic and the Cambridge Computational Biology Institute and senior member in the Bioinformatics Group in the Department of Pathology. The responsibilities in this post were methods developent, co-supervision of research students and junior staff, analytical contributions to research grants in computatopnal biology, data analysis and statistical consulting. My methodological research interest is to provide fully probabilistic models to improve analysis of microarray experiments focusing on aspects how to combine information within multi experiment studies of different biological systems. Visiting Research Fellow in Prof. MacKay’s Inference Group. 09/2000–09/2003 University of Oxford Postdoctoral Research Fellow Oxford, UK Senior member in the Pattern Analysis and Machine Learning Research Group working on Oxford’s BCI project; Collaboration with Research Dept. of Putney Hospital for Neuro-Disabilities; Data analysis for evaluating different cognitive tasks; Implementation and evaluation of an adaptive BCI; Research on Bayesian latent space models, fixed form variational methods and model selection for feature subset selection (work in progress); Implementation of prototypes in MatLab and C++; Writing of project reports for funding agency Tutoring for the Department of Eng. Sci. and for Pembroke College for computer science and electrical engineering courses (Procedural Programming, Functional Programming, Algorithm Design, Numerical Methods for PDE’s, Pattern Recognition, Electrical Machines, Curriculum vitae Dr. Peter Sykacek 1 Communication, Electromagnetic Field, Computer Architecture) Supervising student & visiting scientist projects for the BCI and bioinformatics problems 03/1996–08/2003 Austrian Research Institute for AI Postgraduate Research Assistant Vienna, AT Researcher on two European sleep analysis projects; Collaboration with 16 European partners; Managing responsibility for the feature subset selection task and for the implementation of the prototype sleep analyser; Research on Bayesian classification and time series models; Implementations in MatLab and C++; Major contribution to a funded research proposal; Writing of project reports for funding agencies; Supervising student projects Lecturer at the University of Vienna (03/2000–06/2000) with a course on Bayesian networks 03/1995–02/1996 Systems Engineer Frequentis Vienna, AT Software development for air traffic control equipment; Software engineering in an ISO 9000 certified environment according to the German V-model; Implementations in C in a real time environment Technical liaison with project partners and customers 09/1990–03/1999 Software Trainer Wifi-Wien Vienna, AT Software trainer (part time) for programming languages (Pascal, C, C++), operating systems (all MS platforms) and databases 06/1988–03/1989 Software Engineer MC-Software Vienna, AT Software engineer in a cash register project; Development in Pascal Liaison with customer during first roll out Education 10/1996–06/2000 Technical University of Vienna Vienna, AT PhD in Computer Science / AI (part time) Thesis “Bayesian Inference for Reliable Biomedical Signal Processing” Graduated with a 1.0 equivalent degree (with distinction) Since 1997 various institutions Three seminars at the Isaac Newton Institute for Mathematical Sciences, Cambridge UK About two visits to scientific conferences and workshops per year 10/1984–03/1995 Technical University of Vienna Vienna, AT Dipl Ing. (MSc.) in Electrical Engineering / Control (part time) Thesis “Parameter Optimisation for Fuzzy Control” (in German) Graduated with a 2.1 equivalent degree. Skills Teaching About 9 years of experience teaching computer science subjects in a further Curriculum vitae Dr. Peter Sykacek 2 education institution About 3 years of teaching experience on academic level in German and English Experienced supervisor of student projects and thesis research Research & Project Development About 10 years of research experience in Bayesian methods for machine learning and signal processing Writing of research proposals and project reports for funding agencies Industrial experience Task management skills Interests Organising scientific workshops and seminars; Peer reviewing for several European research councils, for scientific journals and conferences; Sports (running, hiking, experienced ski instructor, wind surfing) Personal Date of Birth: 12/05/65 Marital status: married, one child Publications1 Journal publications (Sykacek et al. 2005) P. Sykacek, R. Furlong and G. Micklem. A Friendly Statistics Package for Microarray Analysis, In Bioinformatics, 21(21). 4096-4070, 2005. (Mukherjee et al. 2004) S.N. Mukherjee, P. Sykacek, S.J. Roberts and S.J. Gurr. Gene Ranking Using Bootstrapped P-values. In ACM SIGKDD Explorations, Volume 5, Issue 2, Special Issue on Microarray Data Mining, 2004. (Sykacek et al. 2004) P. Sykacek, S. J. Roberts and M. Stokes. Adaptive BCI based on variational Bayesian Kalman filtering: an empirical evaluation. In IEEE Trans. Biomedical Engineering, 51(5). 719-729, 2004. (Curran et al. 2004) E. Curran, P. Sykacek, M. Stokes, S. J. Roberts, W. Penny, I. Johnsrude and A. M. Owen. Cognitive tasks for driving a Brain Compute Interfacing System. In IEEE Trans. Neural Systems and Rehabilitation Engineering, 12 (1): 48-57, 2004. (Sykacek et al. 2003) P. Sykacek, S. J. Roberts, M. Stokes, E. Curran, M. Gibbs and L. Pickup. Probabilistic methods in BCI research. In IEEE Trans. Neural Systems and Rehabilitation Engineering, pages 192-195, 2003. (Sykacek et al. 2002) P. Sykacek, G. Dorffner, P. Rappelsberger and J. Zeitlhofer. Improving bio-signal processing through modelling uncertainty: Bayes vs. nonBayes in sleep staging. Applied Artificial Intelligence, 16(5):395421,2002. (Flexer et al. 2002) A. Flexer, G. Dorffner, P. Sykacek, I. Rezek. An automatic, continuous and probabilistic sleep stager based on a hidden Markov model. Applied Artificial Intelligence, 16(3):199-207,2002. (Rappelsberger et al. 2001) P. Rappelsberger, E. Trenker, C. Rothman, G. Gruber, P. Sykacek, S. 1 Many preprints are available at http://www.sykacek.net/pubs.html. Curriculum vitae Dr. Peter Sykacek 3 Roberts, G. Klösch, J. Zeitlhofer, P. Anderer, B. Saletu, A. Schlögl, A. Värri, B. Kemp, T. Penzel, W. M. Herrmann, J. Hasan, M. J. Barbanoj, D. Kunz, G. Dorffner. Das Projekt SIESTA. in Klinische Neurophysiologie,32(2): pages 76-88, 2001. (Rezek et al. 2000b) I. Rezek, P. Sykacek and S. Roberts. Learning interaction dynamics with coupled hidden Markov models. in IEE special issue proceedings science measurement and technology, Vol. 147(5), pages 345-350, 2000. Peer reviewed conference publications and book chapters (Sykacek et al. 2004) P. Sykacek, I. Rezek and S. J. Roberts. Bayes Consistent Classification of EEG Data by Approximate Marginalisation. In S. J. Roberts and R. Dybowski editors Applications of Probabilistic Models for Medical Informatics and Bioinformatics, to appear, Springer Verlag, 2004. (Sykacek & Roberts 2003) P. Sykacek and S. J. Roberts. Adaptive classification by variational Kalman filtering. In S.Thrun, S. Becker and K. Obermayer, editors, Advances in Neural Information Processing Systems 15, pages 737744, MIT press, 2003. (Rezek et al. 2002) I. Rezek, S. J. Roberts and P. Sykacek,(2002). Ensemble Coupled Hidden Markov Models for Joint Characterisation of Dynamic Signals. Ninth International Workshop on Artificial Intelligence and Statistics 2003. (Sykacek & Roberts 2002) P. Sykacek and S. J. Roberts. Bayesian time series classification. In T. G. Dietterich, S. Becker and Z. Ghahramani, editors, Advances in Neural Information Processing Systems 14, 937-944, MIT press, 2002. (Sykacek et al. 2001) P. Sykacek, S. J. Roberts, I. Rezek, A. Flexer and G. Dorffner. A probabilistic approach to high resolution sleep analysis. In G. Dorffner, K. Hornik and H. Bischof, editors, Proceedings of the International Conference on Neural Networks (ICANN), pages 617-624, Springer Verlag, 2001. (Dorffner et al. 2000) G. Dorffner, P. Sykacek and C. Schittenkopf. Modelling Uncertainty in Biomedical Applications of Neural Networks. in Proceedings of Artificial Neural Networks in Medicine and Biology 1, Göteborg, Sweden, pages 18-26, Springer Verlag, 2000. (Flexer et al. 2000) A. Flexer, P. Sykacek, I. Rezek and G. Dorffner. Using hidden Markov models to build an automatic, continuous and probabilistic stager, in S. I. Amari etal., Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks, IJCNN 2000, Como Italy, IEEE Computer Society, Vol. III, 627-631, 2000. (Rezek et al. 2000a) I. Rezek, P. Sykacek and S. Roberts. Coupled hidden Markov models for biosignal interaction modelling. in Proceedings of Medsip-2000, International Conference on Advances in Medical Signal and Information Processing, pages 672-679, 2000. (Sykacek 2000) P. Sykacek.. On input selection with reversible jump Markov chain Monte Carlo sampling. in S. A. Solla and T. K. Leen and K. R. Müller editors, Advances in Neural Information Processing Systems 12, pages 638-644, MIT press, 2000. Curriculum vitae Dr. Peter Sykacek 4 (Sykacek 1998) P. Sykacek. Outliers and Bayesian Inference. in M. Heiss editor, Proceedings of NC 98 Vienna, pages 973-978, 1998. (Cristianini et al. 1998) N. Cristianini, J. Shawe-Taylor and P. Sykacek. Bayesian Classifiers are Large Margin Hyperplanes in a Hilbert Space. in Proc. ICML 98, pages 109-117, 1998. (Sykacek et al. 1998) P. Sykacek, G. Dorffner, P. Rappelsberger and J. Zeitlhofer. Experiences with Bayesian learning in a real world application. in M. I. Jordan and M. J. Kearns and S. Solla editors, Advances in Neural Information Processing Systems 10, pages 964-970, 1998. (Sykacek 1997) P. Sykacek. Equivalent error bars for neural network classifiers trained by Bayesian inference. in Proceedings of the European Symposium on Artificial Neural Networks (Bruges, 1997), pages 121-126, 1997. Curriculum vitae Dr. Peter Sykacek 5