Joelle Pineau's webpage at McGill

News and points of interest:

• Our survey of online POMDP methods:
  • S. Ross, J. Pineau, S. Paquet, B. Chaib-draa. "Online Planning Algorihtms for POMDPs". Journal of Artificial Intelligence Research (JAIR). Volume 32, pages 663-704. 2008. [.pdf]

• Recent papers on our intelligent wheelchair:
  • A. Atrash, R. Kaplow, J. Villemure, R. West, H. Yamani, J. Pineau. "Development and Validation of a Robust Interface for Improved Human-Robot Interaction". International Journal of Social Robotics. 2009. [.pdf]
    
  • W. Honore, A. Atrash, P. Boucher, R. Kaplow, S. Kelouwani, H. Nguyen, J. Villemure, R. West, F. Routhier, P. Stone, C. Dufour, J.-P. Dussault, D. Rock, P. Cohen, L. Demers, R. Forget, J. Pineau. "Human-Oriented Design and Initial Validation of an Intelligent Powered Wheelchair". RESNA Annual Conference. 2010. [.html]

• New ideas on extracting semantic information from Wikipedia:
  • R. West, D. Precup, J. Pineau. "Automatically Suggesting Topics for Augmenting Text Documents". The 19th ACM Conference on Information and Knowledge Management (CIKM). 2010. [.pdf]
    
  • R. West, D. Precup, J. Pineau. "Completing Wikipedia's Hyperlink Structure through Dimensionality Reduction". The 18th ACM Conference on Information and Knowledge Management (CIKM). 2009. [.pdf]
    
  • R. West, J. Pineau and D. Precup. "Wikispeedia: An Online Game for Inferring Semantic Distances between Concepts". International Joint Conferences on Artificial Intelligence (IJCAI). 2009. [.pdf]

• Some of our papers on Bayesian Reinforcement Learning:
  • S. Ross & J. Pineau. "Model-Based Bayesian Reinforcement Learning in Large Structured Domains". Uncertainty in Artificial Intelligence (UAI). 2008.[.pdf]
    
  • S. Ross, B. Chaib-draa, & J. Pineau. "Bayesian Reinforcement Learning in Continuous POMDPs with Application to Robot Navigation". ICRA 2008. [.pdf]
    
  • S. Ross, B. Chaib-draa, & J. Pineau. "Bayes-Adaptive POMDPs". NIPS 2007. [.pdf]

• I recently co-organized the ICAPS'10 POMDP Practitioners Workshop: solving real-world POMDP problems (in collaboration with Matthijs Spaan, Pascal Poupart, Guy Shani and Jason Williams).

Research interests:

• Planning, learning and decision-making
  • Learning and representation of probabilistic systems
  • Reinforcement learning, MDPs, POMDPs
  • Machine learning

• Mobile robotics, human-robot interaction.
  • Design of an intelligent robotic wheelchair (Smartwheeler project)
  • Large-scale dialogue management

• Adaptive treatment design
  • Automated learning of adaptive treatment strategies for chronic depression
  • Computational modelling and adaptive treatment of epilepsy
  • Preference elicitation in medical decision-making

Current projects are described in more detail on my Research page.