|
|
|
|
|
• By Date • By Publication Type • By Research Category • By Author (Currently at CMU) • By Author (Past and Current CORAL Contributors) •
William Uther and Manuela Veloso. TTree: Tree-Based State Generalization with Temporally Abstract Actions. In Proceedings of SARA-2002, Edmonton, Canada, August 2002.
(unavailable)
@Inproceedings(02sara-ttree,
Author = "William Uther and Manuela Veloso",
Title = "T{T}ree: {T}ree-Based State Generalization with
Temporally Abstract Actions",
BookTitle = "Proceedings of SARA-2002",
Month = "August",
Year = 2002,
Address = "Edmonton, Canada",
Annote = "In this paper we describe the Trajectory Tree or
TTree algorithm. TTree takes a set of policies and
pieces them together to solve a Semi-Markov Decision
Problem (SMDP). The algorithm uses a learned tree
based discretization of the state space as an
abstract state description and both supplied and
generated policies as temporally abstract
actions. It uses a generative model of the world to
sample the transition function for the abstract
SMDP. TTree then finds a policy for the abstract
SMDP. In this paper we present the algorithm and
some detailed examples of its execution. Furthermore
we present empirical comparisons to other SMDP
algorithms showing the effectiveness of our
algorithm.",
bib2html_pubtype ={Refereed Conference},
bib2html_rescat ={Multiagent Learning}
)
Generated by bib2html.pl (written by Patrick Riley ) on Tue Oct 09, 2007 00:00:14
|
|
 | |||
| [home] [news] [people] [robosoccer] [projects] [multimedia] [download] [publications] [contact us] | |||