Program (will be updated shortly)

Monday, April 25


Morning: 


9:00 - 9:45 Doina Precup Reinforcement Learning in the Presence of Humans
9:45 - 10:00 Break
10:00 - 10:30 Bei Peng A Need for Speed: Adapting Agent Action Speed to Improve Task Learning from Non-Expert Humans
10:30 - 11:15
Matt Taylor
Dude, Where's My Bias? Improving RL with Human Demonstrations
11:15 - 11:30 Break
11:30 - 12:00 Ann Edwards Machine Learning and Unlearning to Autonomously Switch Between Robotic Functions
12:00 - 12:30 Harm van Seijen Language Understanding - an RL perspective


Evening:

7:00 - 8:00 Large Group Discussion: What do we mean by "task"? (moderated by Matt Taylor)
8:00 - 9:00 Break-out Discussions

Tuesday, April 26

Morning: 

9:00 - 9:45John O'DohertyNeural mechanisms underlying learning and inference in a social context
9:45 - 10:00Break
10:00 - 10:45Jochen TrieschTowards a computational model of speech acquisition in human infants
10:45 - 11:30Elliot LudvigOther People’s Money: Learning and Deciding in a social context.
11:30 - 11:45Break
11:45 - 12:15
Joseph Modayil
AlphaGo: Thoughts on Human Interaction
12:15 - 12:45Vivek VeeriahUsing facial expressions to train adaptive interfaces
Evening: 
7:00 - 8:00Large Group Discussion: Biology and computation of social learning (moderated by John O'Doherty and Elliot Ludvig)
8:00 - 9:00
Break-out Discussions
Wednesday, April 27

Morning: 

9:00 - 9:30Pierre-Luc BaconThe good, the bad and the discovery: the specification problem of options discovery
9:30 - 10:00Jake MenasheLearning Hierarchical Models of Teammates and Adversaries
10:00 - 10:15Break
10:15 - 10:45Kory MathewsonRobots learning from human control and feedback
10:45 - 11:15Clement GehringLearning models from demonstrations
11:15 - 11:30Break
11:30 - 12:00Craig SherstanManaging Agent Interactions through Prediction and Confidence
12:00 - 12:30Adam ParkerA Robot's Perspective on Learning to Communicate

Evening:

7:00 - 8:00Large Group Discussion: Implicit v.s. explicit feedback in learning (moderated by Kory Mathewson, with demo)
8:00 - 9:00Break-out Discussions

Thursday, April 28

Morning: 

9:00 - 9:45Patrick PilarskiBuilding Communicative Capital during Human-Machine Collaboration
9:45 - 10:30
Layla El Asri
Rewards in Dialogue Systems
10:30 - 10:45Break
10:45 - 11:15Josh RomoffMulti-Task RL through Conditional Neural Networks (Condnets)
11:15 - 12:00Rich SuttonToward a New View of Action Selection: The Subgoal Keyboard


End Barbados Workshop 2016