Onur Dikmen

Postdoctoral Researcher
Department of Computer Science,
University of Helsinki

                                                                                                    

Research Interests

- Latent Factor Models, Non-negative Matrix Factorization

- Optimization in Markov Random Fields

- Bayesian Statistics, Monte Carlo Methods, Approximate Inference

- Statistical Signal Processing

- Audio Source Modeling and Separation

Publications

     Journal Papers

- O. Dikmen, Z. Yang, and E. Oja, "Learning the Information Divergence," IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014. [www]

- A. Mozeika, O. Dikmen, and J. Piili, "Consistent Inference of a General Model Using the Pseudolikelihood Method," Physical Review E, vol. 90, 010101(R), 2014. [www]

- O. Dikmen and C. Févotte, "Maximum Marginal Likelihood Estimation for Nonnegative Dictionary Learning in the Gamma-Poisson Model," IEEE Transactions on Signal Processing, vol. 60 (10) , pp. 5163-5175, 2012. [www]

- O. Dikmen and A. T. Cemgil, "Gamma Markov Random Fields for Audio Source Modelling," IEEE Transactions on Audio, Speech, and Language Processing, vol. 18 (3) , pp. 589-601, 2010. [www]

- O. T. Yıldız and O. Dikmen, "Parallel Univariate Decision Trees," Pattern Recognition Letters, vol. 28, no. 7, pp. 825-832, 2007.

     Conference Proceedings

- A. Mesaros, T. Heittola, O. Dikmen, T. Virtanen, "Sound event detection in real life recordings using coupled matrix factorization of spectral representations and class activity annotations," International Conference on Acoustics, Speech and Signal Processing (ICASSP'15), (accepted).

- O. Dikmen and A. T. Cemgil, "Sequential Monte Carlo Samplers for Marginal Likelihood Computation in Multiplicative Exponential Noise Models," in IEEE Statistical Signal Processing Workshop (SSP'14), Gold Coast, Australia, 2014. [www]

- O. Dikmen and A. Mesaros, "Sound Event Detection Using Non-negative Dictionaries Learned From Annotated Overlapping Events," in IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA'13), New Paltz, NY, 2013. [www]

- Z. Yang, T. Hao, O. Dikmen, X. Chen, E. Oja, "Clustering by Nonnegative Matrix Factorization Using Graph Random Walk," in Advances in Neural Information Processing Systems 25 (NIPS'12), Lake Tahoe, USA, 2012. [www]

- D. Rohde, O. Cappé, O. Dikmen, "The simulated online EM algorithm for latent factor models," in AIP Conference Proceedings 1490, 294, 2012.

- O. Dikmen and C. Févotte, "Nonnegative Dictionary Learning in the Exponential Noise Model for Adaptive Music Signal Representation," in Advances in Neural Information Processing Systems 24 (NIPS'11), Granada, Spain, 2011. [www]

- O. Dikmen and C. Févotte, "Maximum Marginal Likelihood Estimation for Nonnegative Dictionary Learning," in Proc. of International Conference on Acoustics, Speech and Signal Processing (ICASSP'11), Prague, Czech Republic, 2011. [www]

- O. Dikmen and A. T. Cemgil, "Score Matching for Models with Latent Variables," in Proc. of IEEE 19th Signal Processing and Communications Applications Conference (SIU'11), Antalya, Turkey, 2011.

- O. Dikmen and A. T. Cemgil, "Unsupervised Single-Channel Source Separation Using Bayesian NMF," in Proc. of IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA'09), pp. 93-96, New Paltz, NY, 2009. [www]

- O. Dikmen, A. T. Cemgil and L. Akarun, "Gamma Markov Random Fields for Audio Source Modelling," in Proc. of IEEE 17th Signal Processing and Communications Applications Conference (SIU'09), Antalya, Turkey, 2009.

- A. T. Cemgil and O. Dikmen, "Inference and learning in Gamma chains for Bayesian audio processing," in Acoustics'08 Paris 123 (5), 2008.

- A. T. Cemgil and O. Dikmen, "Conjugate Gamma Markov random fields for modelling nonstationary sources," in Proc. of 7th International Conference on Independent Component Analysis and Signal Separation (ICA'07), pp. 697-705, 2007.

- A. T. Cemgil, P. Peeling, O. Dikmen and S. J. Godsill, "Prior structures for time-frequency energy distributions," in Proc. of IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA'07), October 2007.

- O. Dikmen, H. L. Akın and E. Alpaydın, "Estimating Distributions In Genetic Algorithms," in Proc. of ISCIS'03, 2003.

     Technical Reports

- O. Dikmen and A. T. Cemgil, "Inference and parameter estimation in Gamma chains," University of Cambridge, Technical Report, CUED/F-INFENG/TR.596, February 2008. [pdf | www]

- O. T. Yıldız and O. Dikmen, "Parallel Univariate Decision Trees," Technical Report, FBE/CMPE-02/2004-13, Institute of Graduate Studies in Science and Engineering, Bogazici University, 2004.

     Theses

- O. Dikmen, "Bayesian Source Modelling for Single-Channel Audio Separation," Ph.D. Thesis, Bogazici University, 2009.

- O. Dikmen, "Estimating Distributions in Genetic Algorithms," M.S. Thesis, Bogazici University, 2002.

Education

Ph.D. : Bogazici University, Computer Engineering, 2009

M.S. : Bogazici University, Computer Engineering, 2002

B.S. : Bogazici University, Computer Engineering, 2000

Contact

Department of Computer Science, University of Helsinki,
Gustaf Halströminkatu 2b, Exactum, 3rd floor, A342,
Kumpula Campus, FI-00014 Helsinki, Finland.

e-mail : name DOT surname AT helsinki DOT fi