Hi, my name is Abdeslam Boularias. I am a research scientist at the Max Planck Institute for Intelligent Systems in Tübingen. I am in the Machine Learning department, led by Prof. Bernhard Schölkopf. From January 2006 to July 2010, I was a PhD student at Laval University under the supervision of Prof. Brahim Chaib-draa. My thesis focused on planning under uncertainty, reinforcement learning, imitation learning, and multi-agent systems. My current research interests focus on reinforcement learning techniques for robotics. I often collaborate with Prof. Jan Peters.
Here is my CV in PDF.
Contact Information
Address: Abdeslam Boularias
Max Planck Institute for Intelligent Systems
Spemannstrasse 38
72076 Tübingen, Germany
Email: boularias@tuebingen.mpg.de
Phone: +49-7071-601-585
Fax: +49-7071-601-552
News
- Fall 2011: I am co-teaching with Jan Peters and Philipp Hennig a course on autonomous intelligent systems (Autonome Lernsysteme) at TU Darmstadt.
- July 2011: I organized with Brian Ziebart and Jan Peters a workshop at ICML on the New Developments in Imitation Learning.
Referred Conference Papers
- Abdeslam Boularias, Oliver Kroemer and Jan Peters (2011). "Learning Robot Grasping from 3-D Images with Markov Random Fields". Appearing in Proceedings of the 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS'11), San Francisco, CA, USA, 2011.[PDF][video]
- Zhikun Wang, Abdeslam Boularias, Katharina Muelling and Jan Peters (2011). "Balancing Safety and Exploitability in Opponent Modeling". Appearing in Proceedings of the Twenty-Fourth National Conference on Artificial Intelligence (AAAI'11), San Francisco, CA, USA, 2011.[PDF]
- Abdeslam Boularias, Jens Kober and Jan Peters (2011). "Relative Entropy Inverse Reinforcement Learning". Appearing in Proceedings of the International Conference on Artificial Intelligence and Statistics (AISTATS'11), Fort Lauderdale, FL, USA, 2011. Volume 15 of JMLR: W&CP 15.[PDF][poster (PDF)]
- Abdeslam Boularias and Brahim Chaib-draa (2010). "Bootstrapping Apprenticeship Learning". In Advances in Neural Information Processing Systems 24 (NIPS'10), Vancouver, Canada, 2010.[PDF][poster (PDF)]
- Abdeslam Boularias and Brahim Chaib-draa (2010). "Apprenticeship Learning via Soft Local Homomorphisms". In Proceedings of 2010 IEEE International Conference on Robotics and Automation (ICRA'10), Anchorage, USA, 2010.[PDF]
- Abdeslam Boularias and Brahim Chaib-draa (2009). "Predictive Representations for Policy Gradient in POMDPs". In Proceedings of the Twenty-sixth International Conference on Machine Learning (ICML'09), Montreal, Canada, 2009.[PDF][poster (PDF)]
- Abdeslam Boularias and Brahim Chaib-draa (2008). "Exact Dynamic Programming for Decentralized POMDPs with Lossless Policy Compression". In Proceedings of the International Conference on Automated Planning and Scheduling (ICAPS'08), Sydney, Australia, 2008.[PDF][poster (PDF)]
- Abdeslam Boularias, Masoumeh Izadi and Brahim Chaib-draa (2008). "Prediction-directed Compression of POMDPs". In Proceedings of the International Conference on Machine Learning and Applications (ICMLA'08), San Diego, CA, USA, 2008.[PDF]
- Abdeslam Boularias (2008). " A Predictive Model for Imitation Learning in Partially Observable Environments". In Proceedings of the International Conference on Machine Learning and Applications (ICMLA'08), San Diego, CA, USA, 2008.[PDF]
- Abdeslam Boularias, Masoumeh Izadi and Brahim Chaib-draa (2008). "State Space Compression with Predictive Representations". In Proceedings of 21st International Florida Artificial Intelligence Research Society Conference (FLAIRS'08), Coconut Grove, FL, USA, 2008.[PDF]
- Andriy Burkov, Abdeslam Boularias and Brahim Chaib-draa (2007). "Competition and Coordination in Stochastic Games". In Proceedings of the 2007 Twentieth Canadian Conference on Artificial Intelligence (Canadian AI'07), Montreal, Canada, May 28-30, 2007.[PDF]
Referred Workshop Papers
- Abdeslam Boularias, Hamid R. Chinaei and Brahim Chaib-draa (2010). "Learning the Reward Model of Dialogue POMDPs". In NIPS'10 Workshop on Machine Learning for Assistive Technology (MLAT-2010) , Whistler, Canada, 2010.[PDF]
- Abdeslam Boularias and Brahim Chaib-draa (2009). "Policy Transfer in Apprenticeship Learning". In NIPS'09 Workshop on Transfer Learning for Structured Data , Whistler, Canada, 2009.[poster (PDF)]
- Abdeslam Boularias and Brahim Chaib-draa (2009). "Learning Probabilistic Models via Bayesian Inverse Planning". In NIPS'09 Workshop on Probabilistic Approaches for Robotics and Control , Whistler, Canada, 2009.[poster (PDF)]
- Abdeslam Boularias and Brahim Chaib-draa (2008). "Planning in Decentralized POMDPs with Predictive Policy Representations". In Proceedings of ICAPS'08 Multiagent Planning Workshop (MASPLAN'08), Sydney, Australia, 2008.[PDF]
- Abdeslam Boularias and Brahim Chaib-draa (2007). "Les Représentations Prédictives des États et des Politiques". In Actes des Quatrièmes Journées Francophones Modèles Formels de l'Interaction (MFI'07), Paris, France, May 30-June 1, 2007.[PDF]
Thesis
- Abdeslam Boularias. "Predictive Representations For Sequential Decision Making Under Uncertainty". PhD Thesis. July 2010.[PDF] [ Presentation slides (PDF)]