NOTICE: Registration closes on August 3, 2017. There will be no onsite registration.

 
Workshop
Fri Aug 11th 08:30 AM -- 05:30 PM @ Parkside 1
Reinforcement Learning Workshop
Doina Precup · Balaraman Ravindran · Pierre-Luc Bacon





The workshop will contain presentations of late-breaking reinforcement learning results in all areas of the field, including deep reinforcement learning, exploration, transfer learning and using auxiliary tasks, theoretical result etc, as well as applications of reinforcement learning to various domains. A panel discussion on the most interesting and challenging current research directions will conclude the workshop.

08:45 AM Introductory Remarks
09:00 AM Some experiments with learning hyperparameters, transfer, and multi-task leaning
Balaraman Ravindran
09:30 AM Talk
Chelsea Finn
10:00 AM Break
10:30 AM Achieving Above-Human Performance on Ms. Pac-Man by Reward Decomposition
Harm van Seijen
11:00 AM Explorartion methods for options
Alessandro Lazaric
11:30 AM The effects of memory replay in reinforcement learning
Ruishan Liu
12:00 PM Lunch
01:30 PM Asynchronous data aggregation for training end-to-end visual control networks
Mathew Monfort
01:50 PM On the reproducibility of policy gradient experiments
Riashat Islam
02:10 PM Reinforcement learning for the MALMO Minecraft challenge
Tudor Berariu
02:30 PM On the convergence of Tree backup and other reinforcement learning algorithms
Ahmed Touati
02:50 PM Transfer learning using successor state features
Lucas Lehnert
03:10 PM Coffee Break
03:30 PM Panel Discussion
Balaraman Ravindran, Chelsea Finn, Alessandro Lazaric, Katja Hofmann, Marc Bellemare