neural networks research group
areas
people
projects
demos
publications
software/data
Design Principles for Creating Human-Shapable Agents (2009)
W. Bradley Knox
and
Ian Fasel
and
Peter Stone
In order for learning agents to be useful to non-technical users, it is important to be able to teach agents how to perform new tasks using simple communication methods. We begin this paper by describing a framework we recently developed called Training an Agent Manually via Evaluative Reinforcement (TAMER), which allows a human to train a learning agent by giving simple scalar reinforcement signals while observing the agent perform the task. We then discuss how this work fits into a general taxonomy of methods for human-teachable (HT) agents and argue that the entire field of HT agents could benefit from an increased focus on the human side of teaching interactions. We then propose a set of conjectures about aspects of human teaching behavior that we believe could be incorporated into future work on HT agents.
View:
PDF
,
PS
,
HTML
Citation:
In
AAAI Spring 2009 Symposium on Agents that Learn from Human Teachers
, March 2009.
Bibtex:
@InProceedings{AAAIsymp09-knox, title={Design Principles for Creating Human-Shapable Agents}, author={W. Bradley Knox and Ian Fasel and Peter Stone}, booktitle={AAAI Spring 2009 Symposium on Agents that Learn from Human Teachers}, month={March}, url="http://nn.cs.utexas.edu/?knox:aaaisymp09", year={2009} }
People
Ian Fasel
ianfasel [at] cs utexas edu
W. Bradley Knox
bradknox [at] mit edu
Peter Stone
pstone [at] cs utexas edu
Projects
Teaching an Agent Manually via Evaluative Reinforcement (TAMER)
Since 2008
Demos
Teaching an Agent Manually via Evaluative Reinforcement (TAMER)
W. Bradley Knox and Peter Stone
2009
Areas of Interest
Reinforcement Learning