Timezone: »

Reinforcement Learning Workshop
Doina Precup · Balaraman Ravindran · Pierre-Luc Bacon

Thu Aug 10 03:30 PM -- 12:30 AM (PDT) @ Parkside 1
Event URL: https://rllabmcgill.github.io/icml2017-rlworkshop/ »

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.

Thu 3:45 p.m. - 4:00 p.m.
Introductory Remarks (Talk)
Thu 4:00 p.m. - 4:30 p.m.
Some experiments with learning hyperparameters, transfer, and multi-task leaning (Talk)
Balaraman Ravindran
Thu 4:30 p.m. - 5:00 p.m.
Chelsea Finn
Thu 5:00 p.m. - 5:30 p.m.
Thu 5:30 p.m. - 6:00 p.m.
Achieving Above-Human Performance on Ms. Pac-Man by Reward Decomposition (Talk)
Harm van Seijen
Thu 6:00 p.m. - 6:30 p.m.
Explorartion methods for options (Talk)
Alessandro Lazaric
Thu 6:30 p.m. - 7:00 p.m.
The effects of memory replay in reinforcement learning (Talk)
Ruishan Liu
Thu 7:00 p.m. - 8:30 p.m.
Lunch (Break)
Thu 8:30 p.m. - 8:50 p.m.
Asynchronous data aggregation for training end-to-end visual control networks (Talk)
Mathew Monfort
Thu 8:50 p.m. - 9:10 p.m.
On the reproducibility of policy gradient experiments (Talk)
Riashat Islam
Thu 9:10 p.m. - 9:30 p.m.
Reinforcement learning for the MALMO Minecraft challenge (Talk)
Tudor Berariu
Thu 9:30 p.m. - 9:50 p.m.
On the convergence of Tree backup and other reinforcement learning algorithms (Talk)
Ahmed Touati
Thu 9:50 p.m. - 10:10 p.m.
Transfer learning using successor state features (Talk)
Lucas Lehnert
Thu 10:10 p.m. - 10:30 p.m.
Coffee Break (Break)
Thu 10:30 p.m. - 11:30 p.m.
Panel Discussion
Balaraman Ravindran, Chelsea Finn, Alessandro Lazaric, Katja Hofmann, Marc Bellemare

Author Information

Doina Precup (McGill University / DeepMind)
Balaraman Ravindran (Indian Institute of Technology)
Pierre-Luc Bacon (McGill University)

More from the Same Authors