Implicit imitation in multiagent reinforcement learning

Implicit imitation in multiagent reinforcement learning (1999)

Cached

Download Links

by Bob Price , Craig Boutilier

Venue:	IN: PROC. ICML
Citations:	32 - 3 self

Versions

Version History

Metadata Version 2

User correction supplied by mph

Datum	Value	Source
TITLE	Implicit imitation in multiagent reinforcement learning	user correction
AUTHOR NAME	Bob Price	user correction
AUTHOR AFFIL	Department of Computer Science; University of British Columbia	user correction
AUTHOR ADDR	Vancouver, B.C., Canada V6T 1Z4	user correction
AUTHOR NAME	Craig Boutilier	user correction
AUTHOR AFFIL	Department of Computer Science; University of British Columbia	user correction
AUTHOR ADDR	Vancouver, B.C., Canada V6T 1Z4	user correction
ABSTRACT	Imitation is actively being studied as an effective means of learning in multi-agent environments. It allows an agent to learn how to act well (perhaps optimally) by passively observing the actions of cooperative teachers or other more experienced agents its environment. We propose a straightforward imitation mechanism called model extraction that can be integrated easily into standard model-based reinforcement learning algorithms. Roughly, by observing a mentor with similar capabilities, an agent can extract information about its own capabilities in unvisited parts of state space. The extracted information can accelerate learning dramatically. We illustrate the benefits of model extraction by integrating it with prioritized sweeping, and demonstrating improved performance and convergence through observation of single and multiple mentors. Though we make some stringent assumptions regarding observability, possible interactions and common abilities, we briefly comment on extensions of the model that relax these.	user correction
YEAR	1999	INFERENCE
VENUE	IN: PROC. ICML	user correction
VENUE TYPE	CONFERENCE	INFERENCE
PAGES	325--334	INFERENCE
CITATIONS	25 found	ParsCit 1.0