Tutorials & Talks

Tutorials

• • The Space of Online Learning Problems, ECML-PKDD, Porto, Portugal, 11 September 2015.

  • PAC-Bayesian Analysis and Its Applications, ECML-PKDD, Bristol, UK, 2012 (with François Laviolette and John Shawe-Taylor)

  • PAC-Bayesian Analysis in Supervised, Unsupervised, and Reinforcement Learning, ICML, Edinburgh, UK, 2012 (with François Laviolette and John Shawe-Taylor)

Summer Schools

• • Concentration Inequalities, Reinforcement Learning Summer School, Barcelona, Spain, 26 June - 5 July, 2023. 

    • Slides. Video.

  • The Space of Online Learning Problems, Online Learning Summer School, Copenhagen, Denmark, 28 June - 2 July, 2015.

Invited Talks

• • Machine learning tools for optimization of energy sector coupling. Digital Tech Summit, Copenhagen, Denmark, 2022.

  • A Strongly Quasiconvex PAC-Bayesian Bound. (Almost) 50 shades of Bayesian learning: PAC-Bayesian trends and insights, NIPS-2017 workshop, Long Beach, California, USA, 2017.

  • New Algorithms for Ranges of Problems. Theoretical Foundations for Learning from Easy Data workshop, Leiden, The Netherlands, 2016.

  • The Space of Online Learning Problems. Multi-armed bandits workshop, Lancaster, UK, 2016.

  • A New Generation of Online Learning Algorithms. Symposium on Intelligent Systems in Science and Industry, Tübingen, Germany, 2015.

  • Interpolation Between Online Learning Settings. Online Algorithms and Learning workshop, Leiden, The Netherlands, 2014.

  • Adaptation in Online Learning. Large-scale Online Learning and Decision Making workshop (LSOLDM-2014), Cumberland Lodge, UK, 2014.

  • PAC-Bayesian Perspective on The Information Bottleneck Method. Information, Representation and Learning: An international workshop in honor of Prof. Naftali Tishby. Jerusalem, Israel, 2013.

Talks

• • An Improved Parametrization and Analysis of the EXP3++ Algorithm for Stochastic and Adversarial Bandits. Conference on Learning Theory (COLT), Amsterdam, The Netherlands, 2017.

  • One Practical Algorithm for Both Stochastic and Adversarial Bandits. International Conference on Machine Learning (ICML), Beijing, China, 2014.

  • Prediction with Limited Advice and Multiarmed Bandits with Paid Observations. International Conference on Machine Learning (ICML), Beijing, China, 2014.

  • PAC-Bayesian Analysis: A Link Between Inference and Statistical Physics. Workshop on Statistical Physics of Inference and Control Theory, Granada, Spain, 2012.

  • PAC-Bayesian Analysis in Unsupervised Learning. Foundations and New Trends of PAC Bayesian Learning Workshop, London, UK, 2010.

  • PAC-Bayesian Approach to Formulation of Clustering Objectives, "Clustering: Science or Art? Towards Principled Approaches", NIPS-2009 Workshop, Whistler, BC, Canada, 2009.

  • Multi-Classification by Categorical Features via Clustering, 25th International Conference on Machine Learning (ICML), Helsinki, Finland, 2008.