Publications

2024

M. K. Titsias, A. Galashov, A. Rannen-Triki, R. Pascanu, Y. W. Teh and J. Bornschein. Kalman Filter for Online Classification of Non-Stationary Data. 12th International Conference on Learning Representations (ICLR), 2024.

2023

M. K. Titsias. Optimal Preconditioning and Fisher Adaptive Langevin Sampling. Neural Information Processing Systems (NeurIPS), 2023.

A. Alexopoulos, P. Dellaportas and M. K. Titsias. Variance reduction for Metropolis–Hastings samplers. Statistics and Computing 33, 6 (2023).

2022

J. Shi, Y. Zhou, J. Hwang, M. K. Titsias and L. Mackey. Gradient Estimation with Discrete Stein Operators. Neural Information Processing Systems (NeurIPS), 2022. NeurIPS 2022 Outstanding Paper Award

M. K. Titsias and J. Shi. Double Control Variates for Gradient Estimation in Discrete Latent Variable Models. 25rd International Conference on Artificial Intelligence and Statistics (AISTATS), 2022.

S. Sun, D. Calandriello, H. Hu, A. Li and M. K. Titsias. Information-theoretic Online Memory Selection for Continual Learning. International Conference on Learning Representations (ICLR), 2022.

2021

M. Hirt, M. K. Titsias, P. Dellaportas. Entropy-based adaptive Hamiltonian Monte Carlo. Neural Information Processing Systems (NeurIPS), 2021.

M. K. Titsias, F. J. R. Ruiz, S. Nikoloutsopoulos and A. Galashov. Information Theoretic Meta Learning with Gaussian Processes. 37th Conference on Uncertainty in Artificial Intelligence (UAI), 2021.

F. J. R. Ruiz, M. K. Titsias, T. Cemgil and A. Doucet. Unbiased Gradient Estimation for Variational Auto-Encoders using Coupled Markov Chains. 37th Conference on Uncertainty in Artificial Intelligence (UAI), 2021.

A. Panos, P. Dellaportas and M. K. Titsias. Large scale multi-label learning using Gaussian processes. Machine Learning, Volume 110, 965–987, 2021.

2020

M. K. Titsias, J. Sygnowski, Y. Chen. Sequential Changepoint Detection in Neural Networks with Checkpoints. arXiv. 2020.

J. Shi, M. K. Titsias and A. Mnih. Sparse Orthogonal Variational Inference for Gaussian Processes. 23rd International Conference on Artificial Intelligence and Statistics (AISTATS), 2020.

M. K. Titsias, J. Schwarz, A. G de G Matthews, R. Pascanu and Y. W. Teh. Functional regularisation for continual learning with Gaussian processes. International Conference on Learning Representations (ICLR), 2020.

2019

A. B. Dieng, F. J. R. Ruiz, D. M. Blei and M. K. Titsias. Prescribed Generative Adversarial Networks. arXiv. 2019.

M. K. Titsias, P. Dellaportas. Gradient-based Adaptive Markov Chain Monte Carlo. Neural Information Processing Systems (NeurIPS), 2019.

F. J. R. Ruiz, M. K. Titsias. A Contrastive Divergence for Combining Variational Inference and MCMC. International Conference on Machine Learning (ICML), 2018.

M. K. Titsias F. J. R. Ruiz. Unbiased Implicit Variational Inference. 22th International Conference on Artificial Intelligence and Statistics (AISTATS), 2019.

K. Martens, M. K Titsias, C. Yau. Rejection-free Ensemble MCMC with applications to Factorial Hidden Markov Models. 22th International Conference on Artificial Intelligence and Statistics (AISTATS), 2019.

2018

F. J. R. Ruiz, M. K. Titsias, A. B. Dieng, and D. M. Blei. Augment and Reduce: Stochastic Inference for Large Categorical Distributions. International Conference on Machine Learning (ICML), 2018.

M. K. Titsias and O. Papaspiliopoulos. Auxiliary gradient-based sampling algorithms. Journal of the Royal Statistical Society: Series B, Vol 80, Issue 4, Pages 749-767, 2018.

2017

M. K. Titsias. Learning Model Reparametrizations: Implicit Variational Inference by Fitting MCMC Distributions. arXiv, 2017.

T. Rukat, C. C. Holmes, M. K. Titsias, C. Yau. Bayesian Boolean Matrix Factorisation. International Conference on Machine Learning (ICML), 2017.

2016

M. K. Titsias. One-vs-Each Approximation to Softmax for Scalable Estimation of Probabilities. Neural Information Processing Systems (NIPS), 29, 2016.

F. J. R. Ruiz, M. K. Titsias and D. M. Blei. The Generalized Reparameterization Gradient. Neural Information Processing Systems (NIPS), 29, 2016.

M. K. Titsias, and C. Yau. The Hamming Ball Sampler. Journal of the American Statistical Association (JASA), Theory and Methods, Vol 112, Issue 520, 2017.

F. J. R. Ruiz, M. K. Titsias and D. M. Blei. Overdispersed Black-Box Variational Inference. Uncertainty in Artificial Intelligence (UAI), 2016. supplementary.

M. Karaliopoulos, I. Koutsopoulos and M. K. Titsias. First Learn then Earn: Optimizing Mobile Crowdsensing campaigns through data-driven user profiling, Proceedings of ACM International Symposium on Mobile Ad-Hoc Networking and Computing (Mobihoc), 2016.

A. Damianou*, M. K. Titsias* and N. Lawrence. Variational Inference for Latent Variables and Uncertain Inputs in Gaussian Processes. Journal of Machine Learning Research (JMLR), 17(42):1-62, 2016. MATLAB code.

M. K. Titsias, C. C. Holmes and C. Yau. Statistical Inference in Hidden Markov Models using k-segment constraints. Journal of the American Statistical Association (JASA), Theory and Methods, 111(513):200-215, 2016.

*Joint first author.

2015

M. K. Titsias and M. Lazaro-Gredilla. Local Expectation Gradients for Black Box Variational Inference. Neural Information Processing Systems (NIPS), 28, 2015. supplementary, sigmoid belief net code.

R. Bardenet* and M. K. Titsias*. Inference for determinantal point processes without spectral knowledge. Neural Information Processing Systems (NIPS), 28, 2015.

*Joint first author.

2014

M. K. Titsias and C. Yau. Hamming Ball Auxiliary Sampling for Factorial Hidden Markov Models. Neural Information Processing Systems (NIPS), 27, 2014. supplementary.

M. K. Titsias and M. Lazaro-Gredilla. Doubly Stochastic Variational Bayes for non-Conjugate Inference. International Conference on Machine Learning (ICML), Beijing, China, 2014. supplementary, MATLAB code.

2013

M. K. Titsias and M. Lazaro-Gredilla. Variational Inference for Mahalanobis Distance Metrics in Gaussian Process Regression. Neural Information Processing Systems (NIPS), 26, 2013. supplementary, MATLAB code.

R. Clifford, T. Louis, P. Robbe, S. Ackroyd, A. Burns, A. T. Timbs, G. W. Colopy, H. Dreau, F. Sigaux, J. G. Judde, M. Rotger, A. Telenti, Y-L Lin, P. Pasero, J. Maelfait, M. Titsias, D. Cohen, S. J. Henderson, M. Ross, D. Bentley, P. Hillmen, A. Pettitt, J. Rehwinkel, S. J. L. Knight, J. C. Taylor, Y. J. Crow, M. Benkirane, A. Schuh. SAMHD1 is mutated recurrently in chronic lymphocytic leukaemia and is involved in response to DNA damage. Blood, 123(7), 1021-31, 2014.

M. Lazaro-Gredilla, M. K. Titsias, J. Verrelst and G. Camps-Valls. Retrieval of Biophysical Parameters with Heteroscedastic Gaussian Processes. IEEE Geoscience and Remote Sensing Letters, 11(4), 838-842, 2014.

2012

M. K. Titsias, A. Honkela, N. D. Lawrence and M. Rattray. Identifying targets of multiple co-regulating transcription factors from expression time-series by Bayesian model comparison. BMC Systems Biology 6:53 (2012).

A. C. Damianou, C. H. Ek, M. K. Titsias and N. D. Lawrence. Manifold Relevance Determination. International Conference on Machine Learning (ICML), 2012.

2011

M. K. Titsias and M. Lazaro-Gredilla. Spike and Slab Variational Inference for Multi-Task and Multiple Kernel Learning. Neural Information Processing Systems (NIPS), 24, 2012. supplementary, code.

A. C. Damianou, M. K. Titsias and N. D. Lawrence. Variational Gaussian Process Dynamical Systems. Neural Information Processing Systems (NIPS), 24, 2012.

M. Lazaro-Gredilla and M. K. Titsias. Variational Heteroscedastic Gaussian Process Regression. International Conference on Machine Learning (ICML), 2011, Distinguished Paper Award.

M. K. Titsias. Discussion on the paper Riemann manifold Langevin and Hamiltonian Monte Carlo methods, by Girolami and Calderhead. Journal of the Royal Statistical Society, Series B (Statistical Methodology), 73(2):201, 2011.

2010

M. K. Titsias and N. D. Lawrence. Bayesian Gaussian Process Latent Variable Model. 13th International Conference on Artificial Intelligence and Statistics (AISTATS), JMLR: W&CP 9, pp. 844-851, 2010.

M. Alvarez, D. Luengo, M. K. Titsias and N. D. Lawrence. Variational inducing kernels for sparse convolved multiple output Gaussian processes. 13th International Conference on Artificial Intelligence and Statistics (AISTATS), JMLR: W&CP 9, pp. 25-32, 2010.

M. K. Titsias, M. Rattray and N.D. Lawrence. Markov chain Monte Carlo algorithms for Gaussian processes. Chapter to appear in the book “Inference and Learning in Dynamic Models” (Cambridge University Press), edited by Barber, Chiappa and Cemgil.

N. D. Lawrence, M. Rattray, P. Gao and M. K. Titsias. Gaussian processes for missing species in biochemical systems. In N. D. Lawrence, M. Girolami, M. Rattray and G. Sanguinetti (eds) Learning and Inference in Computational Systems Biology, MIT Press, Cambridge, MA.

2009

M. K. Titsias. Variational Learning of Inducing Variables in Sparse Gaussian Processes. 12th International Conference on Artificial Intelligence and Statistics, (AISTATS), JMLR: W&CP 5, pp. 567-574, 2009. technical report.

M. K. Titsias, N.D. Lawrence and M. Rattray. Efficient Sampling for Gaussian Process Inference using Control Variables. Neural Information Processing Systems (NIPS), 2009. supplementary.

2008

M. K. Titsias. The Infinite Gamma-Poisson Feature Model. Neural Information Processing Systems (NIPS), 2008.

2006

C. Constantinopoulos, M. K. Titsias and A. Likas. Bayesian Feature and Model Selection for Gaussian Mixture Models. IEEE Trans. on Pattern Analysis and Machine Intelligence, 28(6), 1013-1018, June 2006.

M. K. Titsias and C. K.I. Williams. Sequentially Learning of Layered Models from Video. In C. S. J. Ponce, M. Herbert and A. Zisserman (Eds.), Proceedings Sicily Workshop on Object Recognition, Sicily 2006.

2005

M. K. Titsias. Unsupervised Learning of Multiple Objects in Images. Ph.D. Thesis, School of Informatics, University of Edinburgh, 2005.

M. Allan, M. K. Titsias and C. K.I. Williams. Fast learning of sprites using invariant features. British Machine Vision Conference, 2005. See videos.

M. K. Titsias and C. K.I. Williams. Unsupervised Learning of Multiple Aspects of Moving Objects from Video. Advances in Informatics, PCI 2005, Volos, Greece, 2005.

M. K. Titsias and C. K. I. Williams. Sequentially Fitting Mixtures Models using an Outlier Component. Technical Report, 2005.

2004

M. K. Titsias and C. K. I. Williams. Fast Unsupervised Greedy Learning of Multiple Objects and Parts from Video. Generative-Model Based Vision Workshop, 2004.

C. K.I. Williams and M. K. Titsias. Greedy Learning of Multiple Objects in Images using Robust Statistics and Factorial Learning. Neural Computation, 16(5), 1039-1062, May 2004.

2003

C. K.I. Williams and M. K. Titsias. Learning About Multiple. Objects in Images: Factorial Learning without Factorial Search. Neural Information Processing Systems (NIPS) 15, 2003.

M. K. Titsias and A. Likas. Class conditional density estimation using mixtures with constrained component sharing. IEEE Trans. on Pattern Analysis and Machine Intelligence, 25(7), 924-928, July 2003.

2000-2003

M. K. Titsias and A. Likas. Mixture of experts classification using a hierarchical mixture model. Neural Computation, 14(9), 2221-2244, September 2002.

M. K. Titsias. The Q function of the EM algorithm for hidden variable structure learning. Technical Report, 2002.

M. K. Titsias and A. Likas. Shared kernel models for class conditional density estimation. IEEE Trans. on Neural Networks, 12(5), 987-997, September 2001.

M. K. Titsias. MSc thesis in Greek. University of Ioannina, Greece, June 2001.

M. K. Titsias and A. Likas. A probabilistic RBF network for classification. IJCNN (Como Italy), July 2000.