Dr. Robert Lieck

Robert Lieck is an expert in machine learning and artificial intelligence and joined the DCML as a postdoctoral researcher in 2018.

Robert studied physics and philosophy at Freie Universität Berlin and did his PhD in the Machine Learning & Robotics Lab (Universität Stuttgart) with Prof. Marc Toussaint, focusing on learning and inference with probabilistic models in mixed discrete-continuous systems.

His research focuses on advanced algorithmic methods for active learning, planning and inference and he strives to understand and model the fundamental cognitive capabilities that are essential for processing music, such as the ability of humans to conceive highly abstract structures while remaining flexible when applying them to the world.

Contact

Phone +41 21 69 34841
WebCall
Postal address EPFL CDH DHI DCML
INN 115 (Bâtiment INN)
Station 14
CH-1015 Lausanne
Office INN 115

Publications

Journal Papers

Lieck, Robert and Marc Toussaint (2016). “Temporally Extended Features in Model-Based Reinforcement Learning with Partial Observability”. en. In: Neurocomputing 192, pp. 49–60.

Stefan K. Saevarsson, Gulshan B. Sharma, Heiko Ramm, Robert Lieck, Carol R. Hutchison, Jason Werle, Sigrun Matthiasdottir, Spencer J. Montgomery, Carolina I. Romeo, Stefan Zachow, and Carolyn Anglin (2013). “Kinematic Di erences Between Gender Speci c and Traditional Knee Implants”. In: The Journal of Arthroplasty 28.9, pp. 1543–1550.

Ho, K. C. T., S. K. Saevarsson, H. Ramm, R. Lieck, S. Zachow, G. B. Sharma, E. L. Rex, S. Amiri, B. C. Y. Wu, A. Leumann, and C. Anglin (2012). “Computed Tomography Analysis of Knee Pose and Geometry before and after Total Knee Arthroplasty”. In: Journal of Biomechanics 45.13, pp. 2215–2221.

Saevarsson, Stefan, Gulshan Sharma, Shahram Amiri, Sigrun Montgomery, Heiko Ramm, Derek Lichti, Robert Lieck, Stefan Zachow, and Carolyn Anglin (2012). “Radiological Method for Measuring Patellofemoral Tracking and Tibiofemoral Kinematics before and after Total Knee Replacement”. In: Bone and Joint Research 1.10, pp. 263–271.

Conference Papers

Langhabel, Jonas, Robert Lieck, Marc Toussaint, and Martin Rohrmeier (2017). “Feature Discovery for Sequential Prediction of Monophonic Music”. In: Proceedings of the 18th International Society for Music Information Retrieval Conference, ISMIR 2017. Suzhou, China.

Lieck, Robert and Marc Toussaint (2015). “Discovering Temporally Extended Features for Reinforcement Learning in Domains with Delayed Causalities”. In: European Symposium on Arti cial Neural Networks, Computational Intelligence and Machine Learning. Presses universitaires de Louvain, p. 183.

Saevarsson, Stefan, Gulshan Sharma, Sigrun Montgomery, Karen Ho, Heiko Ramm, Robert Lieck, Stefan Zachow, Carol Hutchison, Jason Werle, and Carolyn Anglin (2012). “Kinematic Comparison Between Gender Speci c and Traditional Femoral Implants”. In: 67th Canadian Orthopaedic Association (COA) Annual Meeting.

Sharma, Gulshan, Karen Ho, Stefan Saevarsson, Heiko Ramm, Robert Lieck, Stefan Zachow, and Carolyn Anglin (2012). “Knee Pose and Geometry Pre- and Post-Total Knee Arthroplasty Using Computed Tomography”. In: 58th Annual Meeting of the Orthopaedic Research Society (ORS).

Saevarsson, Stefan, Gulshan B. Sharma, Spencer J. Montgomery, Karen Ho, Heiko Ramm, Robert Lieck, Stefan Zachow, and Carolyn Anglin (2011). “Kinematic Comparison Between Gender Speci c and Traditional Femoral Implants”. In: Proceedings of the 11th Alberta Biomedical Engineering (BME) Conference (Poster), p. 80.

Workshop Contributions & Technical Reports

Lieck, Robert, Vien Ngo, and Marc Toussaint (2017). “Exploiting Variance Information in Monte-Carlo Tree Search”. In: ICAPS Workshop on Heuristics and Search for Domain-Independent Planning.

Lieck, Robert and Marc Toussaint (2017). “Active Tree Search”. In: ICAPS Workshop on Planning, Search, and Optimization.

Kulick, Johannes, Robert Lieck, and Marc Toussaint (2016). “Cross-Entropy as a Criterion for Robust Interactive Learning of Latent Properties”. In: NIPS Workshop on the Future of Interactive Learning Machines.

Kulick, Johannes, Robert Lieck, and Marc Toussaint (2014a). Active Learning of Hyperparameters: An Expected Cross Entropy Criterion for Active Model Selection. arXiv.

Kulick, Johannes, Robert Lieck, and Marc Toussaint (2014b). The Advantage of Cross Entropy over Entropy in Iterative Information Gathering. arXiv 1409.7552. University of Stuttgart.

Lieck, Robert (2014). Temporally Extended Features: Modeling Delayed Causalities in Reinforcement Learning. AISTATS (MLSS poster session).