NIPS 2017 - presentations from the Reinforcement Learning, Algorithms, Applications sessions

• Off-policy evaluation for slate recommendation • Robust and Efficient Transfer Learning with Hidden Parameter Markov Decision Processes • Inverse Reward Design • Dynamic Safe Interruptibility for Decentralized Multi-Agent Reinforcement Learning • Unifying PAC and Regret: Uniform PAC Bounds for Episodic Reinforcement Learning • Repeated Inverse Reinforcement Learning • Learning multiple visual domains with residual adapters • Natural value approximators: learning when to trust past estimates • EX2: Exploration with Exemplar Models for Deep Reinforcement Learning • Regret Minimization in MDPs with Options without Prior Knowledge • Successor Features for Transfer in Reinforcement Learning • Overcoming Catastrophic Forgetting by Incremental Moment Matching • Fair Clustering Through Fairlets • Fitting Low-Rank Tensors in Constant Time