Schedule: NIPS-08 Workshop on Model Uncertainty and Risk in RL

Morning

• 07:30-07:40 Welcome
• 07:40-08:20 Invited talk: KWIK Learning: Explicit Uncertainty and Decision Making, Michael Littman
  
• 08:20-08:40 Bias Correction and Confidence Intervals for Fitted Q-iteration, Bibhas Chakraborty, Victor Strecher, Susan Murphy
  
• 08:40-09:00 Risk-Aware Decision Making and Dynamic Programming, Boris Defourny, Damien Ernst, Louis Wehenkel
  
• 09:00-09:15 break
• 09:15-09:40 poster spotlights (11 posters x 2 min per poster)
• Paper posters
  
1. Kalman Temporal Differences: Uncertainty and Value Function Approximation, Matthieu Geist, Gabriel Fricout, Olivier Pietquin
2. Missing Data and Uncertainty in Batch Reinforcement Learning, Daniel J. Lizotte, Lacey Gunter, Eric Laber, Susan A. Murphy
   
3. Error Reducing Sampling in Reinforcement Learning, Bruno Scherrer and Shie Mannor
   
4. Towards Global Reinforcement Learning, Milen Pavlov and Pascal Poupart
   
5. Incorporating External Evidence in Reinforcement Learning via Power Prior Bayesian Analysis, Funlade T. Sunmola and Jeremy L. Wyatt
   
6. Uncertainty Handling in Evolutionary Direct Policy Search, Verena Heidrich-Meisner, Christian Igel
7. The Optimal Unbiased Value Estimator and its Relation to LSTD, Steffen Grünewälder, Klaus Obermayer
• Extended abstract posters
1. Model-based Bayesian Reinforcement Learning with Tree-based State Aggregation, Cosmin Paduraru, Doina Precup, Stephane Ross, Joelle Pineau
   
2. Near-Bayesian Exploration in Polynomial Time,  J. Zico Kolter, Andrew Y. Ng
   
3. How Close is Close Enough? Finding Optimal Policies in PAC-style Reinforcement Learning, Emma Brunskill
   
4. Bayesian Exploration using Gaussian Processes: Fast Convergence via Generalization, Erick Chastain and Rajesh P. N. Rao
   
• 09:40-10:30 panel discussion (benchmarks and challenges) and poster session
  

Afternoon

• 15:30-16:10 Invited talk: How Good is Forced-Exploration in Linear Stochastic Bandits? Csaba Szepesvari
  
• 16:10-16:30 Risk Sensitive Control: Mean-Variance Tradeoffs, Steffen Grünewälder, Aki Naito and Klaus Obermayer
  
• 16:30-16:50 PAC-MDP Reinforcement Learning with Bayesian Priors in Deterministic MDPs, Ali Nouri, Michael Littman and Lihong Li
  
• 16:50-17:05 break
• 17:05-17:25 Regret-based Reward Elicitation for Markov Decision Processes, Kevin Regan and Craig Boutilier
  
• 17:25-17:45 Data Biased Robust Counter Strategies, Michael Johanson, Michael Bowling
  
• 17:45-18:30 panel discussion (models that work and don't work) and poster session