Fast Online Q(λ)

Marco Wiering; Jürgen Schmidhuber

doi:10.1023/A:1007562800292

Fast Online Q(λ)

Marco Wiering, Jürgen Schmidhuber

Research output: Contribution to journal › Article › peer-review

60 Scopus citations

Abstract

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

Original language	English (US)
Pages (from-to)	105-115
Number of pages	11
Journal	Machine Learning
Volume	33
Issue number	1
DOIs	https://doi.org/10.1023/A:1007562800292
State	Published - Jan 1 1998
Externally published	Yes

ASJC Scopus subject areas

Artificial Intelligence
Software

Access to Document

10.1023/A:1007562800292

Cite this

@article{890dc12b9b224da6901d47cb8bb8f18e,

title = "Fast Online Q(λ)",

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

author = "Marco Wiering and J{\"u}rgen Schmidhuber",

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

year = "1998",

month = jan,

day = "1",

doi = "10.1023/A:1007562800292",

language = "English (US)",

volume = "33",

pages = "105--115",

journal = "Machine Learning",

issn = "0885-6125",

publisher = "Springer Netherlands",

number = "1",

}

TY - JOUR

T1 - Fast Online Q(λ)

AU - Wiering, Marco

AU - Schmidhuber, Jürgen

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 update complexity of previous online Q(λ) implementations based on lookup tables is bounded by the size of the state/action space. Our faster algorithm's update complexity is bounded by the number of actions. The method 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 update complexity of previous online Q(λ) implementations based on lookup tables is bounded by the size of the state/action space. Our faster algorithm's update complexity is bounded by the number of actions. The method is based on the observation that Q-value updates may be postponed until they are needed.

UR - http://link.springer.com/10.1023/A:1007562800292

UR - http://www.scopus.com/inward/record.url?scp=0032182997&partnerID=8YFLogxK

U2 - 10.1023/A:1007562800292

DO - 10.1023/A:1007562800292

M3 - Article

SN - 0885-6125

VL - 33

SP - 105

EP - 115

JO - Machine Learning

JF - Machine Learning

IS - 1

ER -

Fast Online Q(λ)

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this