Teaching an Agent Manually via Evaluative Reinforcement (TAMER)
Since 2008
The TAMER project seeks to create agents which can be effectively taught behaviors by lay people using positive and negative feedback signals (akin to "shaping" by reward and punishment in animal training).
http://web.media.mit.edu/~bradknox/TAMER.html
People
W. Bradley Knox Ph.D. Student (Alumni) bradknox@mit.edu
Peter Stone Professor pstone@cs.utexas.edu
Ian Fasel Alumni ianfasel@cs.utexas.edu
Adam Setapen Masters Student (Alumni) asetapen@cs.utexas.edu
Publications (10)
Learning Non-Myopically from Human-Generated Reward 2013
W. Bradley Knox and Peter Stone
How Humans Teach Agents: A New Experimental Perspective 2012
W. Bradley Knox and Brian D. Glass and Bradley C. Love and W. Todd Maddox and Peter Stone
Learning from feedback on actions past and intended 2012
W. Bradley Knox and Cynthia Breazeal and Peter Stone
Learning from Human-Generated Reward 2012
W. Bradley Knox
Reinforcement Learning from Human Reward: Discounting in Episodic Tasks 2012
W. Bradley Knox and Peter Stone
Reinforcement Learning with Human and MDP Reward 2012
W. Bradley Knox and Peter Stone
Understanding Human Teaching Modalities in Reinforcement Learning Environments: A Preliminary Report 2011
W. Bradley Knox and Peter Stone
Combining Manual Feedback with Subsequent MDP Reward Signals for Reinforcement Learning 2010
W. Bradley Knox and Peter Stone
Design Principles for Creating Human-Shapable Agents 2009
W. Bradley Knox and Ian Fasel and Peter Stone
Interactively Shaping Agents via Human Reinforcement: The TAMER Framework 2009
W. Bradley Knox and Peter Stone
Related Areas
Reinforcement Learning Supervised Learning Robotics Cognitive Science Advice-taking Learners Transfer Learning Social Agents
Demos
Teaching an Agent Manually via Evaluative Reinforcement (TAMER)W. Bradley Knox and Peter Stone2009
Labs
Learning Agents