Machine Learning Theory

The amount and complexity of available data is steadily increasing. To make use of this wealth of information, computing systems are needed that turn the data into knowledge. 

Machine learning is about developing the required algorithms and software systems that automatically analyse data for making predictions, categorizations, and recommendations. Machine learning algorithms are already an integral part of today's computing systems – for example in search engines, recommender systems, or biometrical applications – and have reached superhuman performance in some domains. DIKU's research pushes the boundaries and aims at more robust, more efficient, and more widely applicable machine learning techniques.

Machine learning at DIKU

Machine learning is a branch of computer science and applied statistics covering algorithms that improve their performance at a given task based on sample data or experience. The machine learning research at DIKU, the Department of Computer Science at the University of Copenhagen, is concerned with the design and analysis of adaptive systems for pattern recognition and behaviour generation.






Please click here for a full list of Yevgeny's papers.

  • Niklas Thiemann, Christian Igel, Olivier Wintenberger, and Yevgeny Seldin. A Strongly Quasiconvex PAC-Bayesian Bound. In Algorithmic Learning Theory (ALT), 2017
  • Oswin Krause, Dídac R. Arbonès, and Christian Igel. CMA-ES with Optimal Covariance Update and Storage Complexity. In Advances in Neural Information Processing Systems (NIPS), 2016
  • Ürün Dogan, Tobias Glasmachers, and Christian Igel. A Unified View on Multi-class Support Vector Classification. Journal of Machine Learning Research 17(45), 2016
  • Fabian Gieseke, Justin Heinermann, Cosmin Oancea, and Christian Igel. Buffer k-d Trees: Processing Massive Nearest Neighbor Queries on GPUs. JMLR W&CP32 (ICML), 2014
  • Yevgeny Seldin, Peter L. Bartlett, Koby Crammer, and Yasin Abbasi-Yadkori. Prediction with limited advice and multiarmed bandits with paid observations. In JMLR W&CP, 32 (ICML), 2014
  • Yevgeny Seldin and Aleksandrs Slivkins. One practical algorithm for both stochastic and adversarial bandits. In JMLR W&CP, 32 (ICML), 2014
  • Kai Brügge, Asja Fischer, and Christian Igel. The flip-the-state transition operator for restricted Boltzmann machines. Machine Learning13, pp. 53-69, 2013
  • Fabian Gieseke, Christian Igel, and Tapio Pahikkala. Polynomial runtime bounds for fixed-rank unsupervised least-squares classification. JMLR W&CP 29 (ACML), pp. 62-71, 2013
  • Oswin Krause, Asja Fischer, Tobias Glasmachers, and Christian Igel. Approximation properties of DBNs with binary hidden units and real-valued visible units. JMLR W&CP 28 (ICML), pp. 419–426, 2013
  • Ilya Tolstikhin and Yevgeny Seldin. PAC-Bayes-Empirical-Bernstein Inequality. In Advances in Neural Information Processing Systems (NIPS), 2013
  • Kim Steenstrup Pedersen, Kristoffer Stensbo-Smidt, Andrew Zirm, and Christian Igel. Shape Index Descriptors Applied to Texture-Based Galaxy Analysis. International Conference on Computer Vision (ICCV), pp 2440-2447, IEEE Press, 2013
  • Yevgeny Seldin, François Laviolette, Nicolò Cesa-Bianchi, John Shawe-Taylor, and Peter Auer. PAC-Bayesian inequalities for martingales. IEEE Transactions on Information Theory, 58(12), pp. 7086-7093, 2012
  • Asja Fischer and Christian Igel. Bounding the Bias of Contrastive Divergence Learning. Neural Computation23, pp. 664-673, 2011
  • Yevgeny Seldin, Peter Auer, François Laviolette, John Shawe-Taylor, and Ronald Ortner. PAC-Bayesian analysis of contextual bandits. In Advances in Neural Information Processing Systems (NIPS), 2011
  • Tobias Glasmachers and Christian Igel. Maximum Likelihood Model Selection for 1-Norm Soft Margin SVMs with Multiple Parameters. IEEE Transactions on Pattern Analysis and Machine Intelligence 32(8), pp. 1522-1528, 2010 source code
  • Yevgeny Seldin and Naftali Tishby. PAC-Bayesian analysis of co-clustering and beyond. Journal of Machine Learning Research11, pp. 3595−3646, 2010
  • Thorsten Suttorp, Nikolaus Hansen, and Christian Igel. Efficient Covariance Matrix Update for Variable Metric Evolution Strategies. Machine Learning 75, pp. 167-197, 2009 source code
  • Verena Heidrich-Meisner and Christian Igel. Hoeffding and Bernstein Races for Selecting Policies in Evolutionary Direct Policy Search. In L. Bottou and M. Littman, eds.: Proceedings of the International Conference on Machine Learning (ICML 2009), pp. 401-408, 2009
  • Christian Igel, Verena Heidrich-Meisner, and Tobias Glasmachers. Shark. Journal of Machine Learning Research 9, pp. 993-996, 2008 source code
  • Tobias Glasmachers and Christian Igel. Maximum-Gain Working Set Selection for SVMs. Journal of Machine Learning Research 7, pp. 1437-1466, 2006 source code



Name Title Phone E-mail
Gieseke, Fabian Christian Associate Professor   E-mail
Igel, Christian Professor +4535335674 E-mail
Krause, Oswin Associate Professor   E-mail
Seldin, Yevgeny Professor +4530450082 E-mail
Talebi, Sadegh Assistant Professor - Tenure Track +4531539770 E-mail