Sinan Yıldırım

home | work | teaching

Journal

Exact MCMC with differentially private moves, S. Yıldırım, Beyza Ermiş, Statistics and Computing, 2019

A Markov chain Monte Carlo algorithm for Bayesian policy search, Vahid Tavakol, Ahmet Onat and S. Yıldırım, Systems Science and Control Engineering, 2018

Bayesian tracking and parameter learning for non-linear multiple target tracking models, Lan Jiang, Sumeetpal S. Singh, Sinan Yıldırım, IEEE Transactions on Signal Processing, 2015

Parameter estimation in hidden Markov Models with intractable likelihoods using sequential Monte Carlo, S. Yıldırım, S. Singh, T. Dean, A. Jasra, Journal of Computational and Graphical Statistics, 2015

Calibrating the gaussian multi-target tracking model, S.Yıldırım, L. Jiang, S. S. Singh, and T. A. Dean, Statistics and Computing, 2015

An online expectation-maximisation algorithm for changepoint models, S. Yıldırım, S. S. Singh, and A. Doucet, Journal of Computational and Graphical Statistics, 2013

A Bayesian deconvolution approach for receiver function analysis, S. Yıldırım, A. T. Cemgil, M. Aktar, Y. Özakın, and A. Ertüzün, IEEE Transactions in Geoscience and Remote Sensing, 2010

Estimation of time varying AR SαS processes using Gibbs sampling, D. Gençağa, E. E. Kuruoğlu, A. Ertüzün, and S. Yıldırım, Signal Processing, 2008

Conference

MCMC shape sampling for image segmentation with nonparametric shape priors, E. Erdil, S. Yıldırım, T. Taşdizen, M. Çetin, Computer Vision and Pattern Recognition (CVPR), 2016

A new particle filtering algorithm for multiple target tracking with non-linear observations, L. Jiang, S. S. Singh, S.Yıldırım, 17’th International Conference on Information Fusion, 2014

A Monte Carlo expectation-maximisation algorithm for parameter estimation in multiple target tracking, S. Yıldırım, L. Jiang, S. S. Singh, and T. Dean, 15’th International Conference on Information Fusion, 2012

An online expectation-maximisation algorithm for nonnegative matrix factorisation models, S. Yıldırım, A. T. Cemgil, and S. S. Singh, SYSID, 2012

A hybrid method for deconvolution of Bernoulli-Gaussian processes, S. Yıldırım, A. T. Cemgil, and A. Ertüzün, IEEE ICASSP, 2009

Workshop papers

Image segmentation with pseudo-marginal MCMC sampling and nonparametric shape priors, Ertunç Erdil, S. Yıldırım, Tolga Taşdizen, Müjdat Çetin, NIPS Workshop on Advances in Approximate Bayesian Inference, 2017

Bayesian nonnegative matrix factorization as an allocation model, M. Burak Kurutmaz, A. Taylan Cemgil, Umut Şimşekli, S. Yıldırım, NIPS Workshop on Advances in Approximate Bayesian Inference, 2017

Reports

Scalable Monte Carlo inference in state-space models, S. Yıldırım, Christophe Andrieu, Arnaud Doucet, arxiv, 2018

On the utility of Metropolis-Hastings with asymmetric acceptance ratio, Christophe Andrieu, Arnaud Doucet, S. Yıldırım, Nicolas Chopin, in arXiv, 2018

On the Use of Penalty MCMC for Differential Privacy, S. Yıldırım, arxiv, 2016

In progress

Pseudo-marginal MCMC Sampling for image segmentation using nonparametric shape priors, Ertunç Erdil, S. Yıldırım, Tolga Taşdizen, Müjdat Çetin (2018), submitted to IEEE Transactions on Image Processing, 2018

Data sharing via differentially private coupled tensor factorization, Beyza Ermiş, Taylan Cemgil, S. Yıldırım, submitted to The Journal of Machine Learning Research, 2018

Theses

Maximum Likelihood Parameter Estimation in Time Series Models Using Sequential Monte Carlo, PhD, Department of Pure Mathematics and Mathematical Statistics, University of Cambridge, UK, 2013

Bayesian Methods for Deconvolution of Sparse Processes, MSc, Department of Electrical and Electronics Engineering, Boğaziçi University, Turkey, 2009

top

ˆ