Shimon Whiteson

About Me

I am an associate professor, ERC Starting Grant recipient, and VIDI laureate at the Informatics Institute at the University of Amsterdam.  I lead the Autonomous Agents section of the Intelligent Autonomous Systems group.  My research focuses on single- and multi-agent decision-theoretic planning and learning, including reinforcement learning, multi-agent planning, and stochastic optimization.

Current research efforts include multi-objective multi-agent planning, planning in decentralized partially observable Markov decision processes, and reinforcement learning for information retrieval. I am also the project leader and scientific coordinator for the TERESA Project. To find out more about my research, check out my research page or see my publication list.


April 2015: In October I will be moving to the University of Oxford as an associate professor in the Department of Computer Science and as a fellow of St. Catherine’s College.

April 2015: Our paper Bayesian Ranker Comparison Based on Historical User Interactions was accepted for publication in SIGIR-2015.

April 2015: Our paper Point-Based Planning for Multi-Objective POMDPs was accepted for publication in IJCAI-2015.

April 2015: The new website for the TERESA Project on semi-autonomous telepresence robots is now live.

February 2015: Our article Computing Convex Coverage Sets for Faster Multi-Objective Coordination was accepted for publication in the Journal of AI Research.

January 2015: Diederik Roijers and I will present a tutorial on Multi-Objective Decision-Making at IJCAI-15.

December 2014: I have been awarded an ERC Starting Grant by the European Research Council for a project entitled Coevolutionary Policy Search.

November 2014: Together with other members of the TERESA project, I am organizing a workshop at ICRA-2015 on Machine Learning for Social Robotics.

November 2014: Our paper Exploiting Submodular Value Functions for Faster Dynamic Sensor Selection was accepted for publication in AAAI-2015.

October 2014: Our paper MergeRUCB: A Method for Large-Scale Online Ranker Evaluation was accepted for publication in WSDM-2015.