Peter Stone's Selected Publications

Classified by TopicClassified by Publication TypeSorted by DateSorted by First Author Last NameClassified by Funding Source


A Task Specification Language for Bootstrap Learning

Ian Fasel, Michael Quinlan, and Peter Stone. A Task Specification Language for Bootstrap Learning. In AAAI Spring 2009 Symposium on Agents that Learn from Human Teachers, March 2009.
AAAI Spring 2009 Symposium: Agents that Learn from Human Teachers

Download

[PDF]407.2kB  [postscript]1.9MB  

Abstract

Reinforcement learning (RL) is an effective framework for online learning by autonomous agents. Most RL research focuses on domain-independent learning algorithms, requiring an expert human to define the environment (state and action representation) and task to be performed (e.g. start state and reward function) on a case-by-case basis. In this paper, we describe a general language for a teacher to specify sequential decision making tasks to RL agents. The teacher may communicate properties such as start states, reward functions, termination conditions, successful execution traces, task decompositions, and other advice. The learner may then practice and learn the task on its own using any RL algorithm. We demonstrate our language in a simple BlocksWorld example and on the RoboCup soccer keepaway benchmark problem. The language forms the basis of a larger ``Bootstrap Learning'' model for machine learning, a paradigm for incremental development of complete systems through integration of multiple machine learning techniques.

BibTeX Entry

@InProceedings{AAAIsymp09-fasel,
	author="Ian Fasel and Michael Quinlan and Peter Stone",
	title="A Task Specification Language for Bootstrap Learning",
	booktitle="AAAI Spring 2009 Symposium on Agents that Learn from Human Teachers",
	month="March",
	year="2009",
	abstract={Reinforcement learning (RL) is an effective framework for online
		learning by autonomous agents. Most RL research focuses on
		domain-independent learning \emph{algorithms}, requiring an expert
		human to define the \emph{environment} (state and action
		representation) and \emph{task} to be performed (e.g.\ start state and
		reward function) on a case-by-case basis. In this paper, we describe a
		general language for a teacher to specify
		sequential decision making tasks to RL agents.
		The teacher may communicate properties such as
		start states, reward functions, termination conditions, successful
		execution traces, task decompositions, and other advice. The learner
		may then practice and learn the task on its own using any RL
		algorithm. We demonstrate our language in a simple BlocksWorld example and on the RoboCup
		soccer keepaway benchmark problem. The language forms the basis of a larger ``Bootstrap Learning''
		model for machine learning, a paradigm for incremental development of complete systems through
		integration of multiple machine learning techniques.},
	wwwnote={<a href="http://www.aaai.org/Symposia/Spring/sss09.php">AAAI Spring 2009 Symposium: Agents that Learn from Human Teachers</a>},
}

Generated by bib2html.pl (written by Patrick Riley ) on Mon Sep 22, 2014 22:47:23