The workshop will have a mix of short and longer talks and discussions on relevant topics (see here for the way the program was organized last year). Details of this year's talks to follow. Tuesday: Perception Morning: 09h00-10h00: Rich Sutton: The role of perceptual learning in a reinforcement-learning architecture 10h00-10h30: Thomas Degris: Critterbot:
the Problem of Low Level Control 10h30-10h45: Break 10h45-11h15: Patrick Pilarski: Feature and Variable Selection: Context for Representation Learning in RL 11h15-11h45: Adam White: Feature Selection for Hearts and a Trail Towards Feature Discovery for a Mobile Robot 11h45-12h00: Break 12h00-13h00: Csaba Szepesvari: Towards Creating
Adaptive RL Algorithms Evening: Curiosity 19h00-19h30: Joseph Modayil: Computational Curiosity 19h30-20h00: Istvan Szita: Optimism or Curiosity? Thoughts about exploration 20h00-20h20:
Break 20h20-22h00: Open discussion Wednesday: Psychology Morning: 09h00-09h20: Elliot Ludvig: Perceptual Learning in Humans and Other Animals 09h20-10h05: Yael Niv: Learning Latent Structures 10h05-10h50: Todd Gureckis: Five Simple Principles of Human Learning That You Should Know About 10h50-11h15: Break 11h15-12h00: Nathaniel Daw: Reinforcement learning in the brain: Beyond reinforcement 12h00-12h30:
Francois Rivest: Learning and the partial observability of
continuous time 12h30-13h00: Phil
Bachman: Feature Extraction from Video Images Evening: 19h00-20h20: Eduardo Alonso & Elliot Ludvig: Discussion on RL in Animal Learning 20h20-20h40:
Break 20h40-22h00: Ozgur Simsek: Simple
heuristics that make us smart + Discussion Thursday: Information Morning: 09h00-10h00: Naftali Tishby: Robust tradeoff between Value and Future Information 10h00-10h30: Keith Bush: Applications of Dynamic Bases to Reinforcement Learning 10h30-10h50: Discussion 10h50-11h10: Break 11h10-11h40: Susan Shortreed: Learning the Similarity Matrix for Spectral Clustering 11h40-12h10: Jordan Frank: Activity and Gait recognition using time-delay embeddings 12h10-12h30: Discussion Evening: 19h00-22h00: Self-directed Break-Out Groups Friday: Discovery Morning: 09h00-10h00: Pascal Poupart: Hierarchy Discovery in Partially Observable Domains 10h00-10h30: Brad Knox: Combining Manual Feedback with Subsequent MDP Reward Signals for Reinforcement Learning 10h30-11h00: Break 11h00-11h30: Gheorghe Comanici: Policy Switching towards transferring behavioral knowledge 11h30-12h00: General Discussion End of
Barbados Workshop 2010 |
 |
|
