Partitioning input space for reinforcement learning for control

D. F. Hougen; M. Gini; J. Slagle

doi:10.1109/ICNN.1997.616117

Partitioning input space for reinforcement learning for control

D. F. Hougen, M. Gini, J. Slagle

Computer Science and Engineering

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

8 Scopus citations

Abstract

This paper considers the effect of input-space partitioning on reinforcement learning for control. In many such learning systems, the input space is partitioned by the system designer. However, input-space partitioning could be learned. Our objective is to compare learned and fixed input-space partitionings in terms of the overall system learning speed and proficiency achieved. We present a system for unsupervised control-learning in temporal domains with results for both fixed and learned input-space partitionings. The trailer-backing task is used as an example problem.

Original language	English (US)
Title of host publication	1997 IEEE International Conference on Neural Networks, ICNN 1997
Pages	755-760
Number of pages	6
DOIs	https://doi.org/10.1109/ICNN.1997.616117
State	Published - Dec 1 1997
Event	1997 IEEE International Conference on Neural Networks, ICNN 1997 - Houston, TX, United States Duration: Jun 9 1997 → Jun 12 1997

Publication series

Name	IEEE International Conference on Neural Networks - Conference Proceedings
Volume	2
ISSN (Print)	1098-7576

Other

Other	1997 IEEE International Conference on Neural Networks, ICNN 1997
Country/Territory	United States
City	Houston, TX
Period	6/9/97 → 6/12/97

Access

10.1109/ICNN.1997.616117

OpenUrl availability

Full text

Cite this

Hougen, DF, Gini, M & Slagle, J 1997, Partitioning input space for reinforcement learning for control. in 1997 IEEE International Conference on Neural Networks, ICNN 1997., 616117, IEEE International Conference on Neural Networks - Conference Proceedings, vol. 2, pp. 755-760, 1997 IEEE International Conference on Neural Networks, ICNN 1997, Houston, TX, United States, 6/9/97. https://doi.org/10.1109/ICNN.1997.616117

@inproceedings{db84ed485ee04ab88927564a14b21f01,

title = "Partitioning input space for reinforcement learning for control",

abstract = "This paper considers the effect of input-space partitioning on reinforcement learning for control. In many such learning systems, the input space is partitioned by the system designer. However, input-space partitioning could be learned. Our objective is to compare learned and fixed input-space partitionings in terms of the overall system learning speed and proficiency achieved. We present a system for unsupervised control-learning in temporal domains with results for both fixed and learned input-space partitionings. The trailer-backing task is used as an example problem.",

author = "Hougen, {D. F.} and M. Gini and J. Slagle",

year = "1997",

month = dec,

day = "1",

doi = "10.1109/ICNN.1997.616117",

language = "English (US)",

isbn = "0780341228",

series = "IEEE International Conference on Neural Networks - Conference Proceedings",

pages = "755--760",

booktitle = "1997 IEEE International Conference on Neural Networks, ICNN 1997",

note = "1997 IEEE International Conference on Neural Networks, ICNN 1997 ; Conference date: 09-06-1997 Through 12-06-1997",

}

TY - GEN

T1 - Partitioning input space for reinforcement learning for control

AU - Hougen, D. F.

AU - Gini, M.

AU - Slagle, J.

PY - 1997/12/1

Y1 - 1997/12/1

N2 - This paper considers the effect of input-space partitioning on reinforcement learning for control. In many such learning systems, the input space is partitioned by the system designer. However, input-space partitioning could be learned. Our objective is to compare learned and fixed input-space partitionings in terms of the overall system learning speed and proficiency achieved. We present a system for unsupervised control-learning in temporal domains with results for both fixed and learned input-space partitionings. The trailer-backing task is used as an example problem.

AB - This paper considers the effect of input-space partitioning on reinforcement learning for control. In many such learning systems, the input space is partitioned by the system designer. However, input-space partitioning could be learned. Our objective is to compare learned and fixed input-space partitionings in terms of the overall system learning speed and proficiency achieved. We present a system for unsupervised control-learning in temporal domains with results for both fixed and learned input-space partitionings. The trailer-backing task is used as an example problem.

UR - http://www.scopus.com/inward/record.url?scp=0030691689&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030691689&partnerID=8YFLogxK

U2 - 10.1109/ICNN.1997.616117

DO - 10.1109/ICNN.1997.616117

M3 - Conference contribution

AN - SCOPUS:0030691689

SN - 0780341228

SN - 9780780341227

T3 - IEEE International Conference on Neural Networks - Conference Proceedings

SP - 755

EP - 760

BT - 1997 IEEE International Conference on Neural Networks, ICNN 1997

T2 - 1997 IEEE International Conference on Neural Networks, ICNN 1997

Y2 - 9 June 1997 through 12 June 1997

ER -

Partitioning input space for reinforcement learning for control

Abstract

Publication series

Other

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this