Papers

Best Paper

The best paper award is generously sponsored by Intel

Best paper: Learning with options that terminate off-policy, Anna Harutyunyan, Peter Vrancx, Pierre-Luc Bacon, Doina Precup, and Ann Nowe

Award committee: Andrew Barto (UMass), Tom Schaul (DeepMind), Roy Fox (UCB)

The award will be presented at the workshop by Milena Marinova, Senior Director of AI Solutions, Intel AIPG

Accepted Papers

We accepted 20 papers of 30 excellent submissions:

• Toward Good Abstractions for Lifelong Learning, David Abel, Dilip Arumugam, Lucas Lehnert, and Michael Littman
• Landmark Options Via Reflection (LOVR) in Multi-task Lifelong Reinforcement Learning, Nicholas Denis and Maia Fraser
• Combining intrinsic motivation and hierarchical reinforcement learning, Maria K. Eckstein and Anne GE Collins
• Empirical Evaluation of Optimism with Options, Ronan Fruit, Matteo Pirotta, Alessandro Lazaric, and Emma Brunskill
• Importance Sampled Option-Critic for More Sample Efficient Reinforcement Learning, Karan Goel and Emma Brunskill
• Optimal Hierarchical Policy Extraction From Noisy Imperfect Demonstrations, Karan Goel, Tong Mu, and Emma Brunskill
• Reinforcement Learning for Recursive Markov Decision Processes, Joshua Gruenstein
• Learning with options that terminate off-policy, Anna Harutyunyan, Peter Vrancx, Pierre-Luc Bacon, Doina Precup, and Ann Nowe
• Learnings Options End-to-End for Continuous Action Tasks, Martin Klissarov, Pierre-Luc Bacon, Jean Harb, and Doina Precup
• Effective Master-Slave Communication On A Multi-Agent Deep Reinforcement Learning System, Xiangyu Kong, Bo Xin, Fangchen Liu, and Yizhou Wang
• Feudal Learning for Large Discrete Action Spaces with Recursive Substructure [appx.], Aviral Kumar, Kevin Swersky, and Geoffrey Hinton
• Federated Control with Hierarchical Multi-Agent Deep Reinforcement Learning, Saurabh Kumar, Pararth Shah, Dilek Hakkani-Tur, and Larry Heck
• Characterizing optimal hierarchical policy inference on graphs via non-equilibrium thermodynamics, Daniel McNamee
• The Eigenoption-Critic Framework, Miao Liu, Marlos C. Machado, Gerald Tesauro, and Murray Campbell
• Crossmodal Attentive Skill Learner, Shayegan Omidshafiei, Dong-Ki Kim, Jason Pazis, and Jonathan How
• Reasoning for reinforcement learning, Silviu Pitis
• Hindsight policy gradients, Paulo Rauber, Filipe Mutz, and Juergen Schmidhuber
• Deep Abstract Q-Networks, Melrose Roderick, Christopher Grimm, and Stefanie Tellex
• Hierarchical and Interpretable Skill Acquisition in Multi-task Reinforcement Learning, Tianmin Shu, Caiming Xiong, and Richard Socher
• A Demon Control Architecture with Off-Policy Learning and Flexible Behavior Policy, Shangtong Zhang and Richard Sutton