Dr. W. Bradley Knox
W. Bradley Knox, Ph.D.
recently completed a postdoctoral researcher position at the MIT Media
Lab. His research interests span machine learning, human-robot
interaction, and psychology, especially machine learning algorithms that
learn through human interaction.
Brad earned his Ph.D. in computer science at the University of Texas at Austin and his BS in psychology and a minor in philosophy from Texas A&M University. At UT-Austin, he was an NSF Graduate Research Fellow from 2008 to 2011 and a member of teams that won one Robocup competition and placed 2nd in two others.
Brad’s research won the Best Paper Award at the International Conference on Social Robotics (ICSR) in 2013 and the Pragnesh Jay Modi Best Student Paper Award at the International Conference on Autonomous Agents and Multiagent Systems (AAMAS) in 2010. His dissertation “Learning from Human-Generated Reward” — advised by Peter Stone — received the Bert Kay Dissertation Award from his doctoral department and a runner-up distinction for the IFAAMAS-12 Victor Lesser Distinguished Dissertation Award. He was named to IEEE’s AI 10 to Watch in 2013.
He conducts research on intersections of (1) the control of robots and other agents, (2) machine learning (reinforcement learning in particular), (3) human-computer interaction, and (4) computational models of human behavior for cognitive science research. He’s primarily interested in human interaction with machine learning algorithms, especially when the human fills a teaching role.
Brad’s papers include Interactively shaping agents via human reinforcement: The TAMER framework, Combining manual feedback with subsequent MDP reward signals for reinforcement learning, Know thine enemy: A champion RoboCup coach agent, TAMER: Training an agent manually via evaluative reinforcement, Reinforcement learning from simultaneous human and MDP reward, How humans teach agents: A new experimental perspective, Design Principles for Creating Human-Shapable Agents, and Learning from Human-Generated Reward. Read the full list of his publications!
Watch Learning from Human-Generated Reward — W. Bradley Knox’s Ph.D. Defense. Watch his YouTube channel. Read his LinkedIn profile.