Speeding up q(λ)-learning

Marco Wiering; Jurgen Schmidhuber

doi:10.1007/bfb0026706

Speeding up q(λ)-learning

Marco Wiering, Jurgen Schmidhuber

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Scopus citations

Abstract

Q(λ)-learning uses TD(λ)-methods to accelerate Q-learning. The worst case complexity for a single update step of previous online Q(λ) implementations based on lookup-tables is bounded by the size of the state/action space. Our faster algorithm's worst case complexity is bounded by the number of actions. The algorithm is based on the observation that Q-value updates may be postponed until they are needed.

Original language	English (US)
Title of host publication	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Publisher	Springer Verlag
Pages	352-363
Number of pages	12
ISBN (Print)	3540644172
DOIs	https://doi.org/10.1007/bfb0026706
State	Published - Jan 1 1998
Externally published	Yes

Bibliographical note

Generated from Scopus record by KAUST IRTS on 2022-09-14

Access to Document

10.1007/bfb0026706

Cite this

@inproceedings{0ac88099b51e4cf5ac90942ee103572f,

title = "Speeding up q(λ)-learning",

abstract = "Q(λ)-learning uses TD(λ)-methods to accelerate Q-learning. The worst case complexity for a single update step of previous online Q(λ) implementations based on lookup-tables is bounded by the size of the state/action space. Our faster algorithm's worst case complexity is bounded by the number of actions. The algorithm is based on the observation that Q-value updates may be postponed until they are needed.",

author = "Marco Wiering and Jurgen Schmidhuber",

note = "Generated from Scopus record by KAUST IRTS on 2022-09-14",

year = "1998",

month = jan,

day = "1",

doi = "10.1007/bfb0026706",

language = "English (US)",

isbn = "3540644172",

pages = "352--363",

booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

address = "Germany",

}

TY - GEN

T1 - Speeding up q(λ)-learning

AU - Wiering, Marco

AU - Schmidhuber, Jurgen

N1 - Generated from Scopus record by KAUST IRTS on 2022-09-14

PY - 1998/1/1

Y1 - 1998/1/1

N2 - Q(λ)-learning uses TD(λ)-methods to accelerate Q-learning. The worst case complexity for a single update step of previous online Q(λ) implementations based on lookup-tables is bounded by the size of the state/action space. Our faster algorithm's worst case complexity is bounded by the number of actions. The algorithm is based on the observation that Q-value updates may be postponed until they are needed.

AB - Q(λ)-learning uses TD(λ)-methods to accelerate Q-learning. The worst case complexity for a single update step of previous online Q(λ) implementations based on lookup-tables is bounded by the size of the state/action space. Our faster algorithm's worst case complexity is bounded by the number of actions. The algorithm is based on the observation that Q-value updates may be postponed until they are needed.

UR - http://link.springer.com/10.1007/BFb0026706

UR - http://www.scopus.com/inward/record.url?scp=84957063167&partnerID=8YFLogxK

U2 - 10.1007/bfb0026706

DO - 10.1007/bfb0026706

M3 - Conference contribution

SN - 3540644172

SP - 352

EP - 363

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

ER -

Speeding up q(λ)-learning

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this