Reinforcement learning algorithms as function optimizers

Ronald J. Williams, Jing Peng

Research output: Contribution to conferencePaper

15 Citations (Scopus)

Abstract

Any nonassociative reinforcement learning algorithm can be viewed as a method for performing function optimization through (possibly noise-corrupted) sampling of function values. A description is given of the results of simulations in which the optima of several deterministic functions studied by D. H. Ackley (Ph.D. Diss., Carnegie-Mellon Univ., 1987) were sought using variants of REINFORCE algorithms. Results obtained for certain of these algorithms compare favorably to the best results found by Ackley.

Original languageEnglish
Pages89-95
Number of pages7
StatePublished - 1 Dec 1989
EventIJCNN International Joint Conference on Neural Networks - Washington, DC, USA
Duration: 18 Jun 198922 Jun 1989

Other

OtherIJCNN International Joint Conference on Neural Networks
CityWashington, DC, USA
Period18/06/8922/06/89

Fingerprint

Reinforcement learning
Learning algorithms
Sampling

Cite this

Williams, R. J., & Peng, J. (1989). Reinforcement learning algorithms as function optimizers. 89-95. Paper presented at IJCNN International Joint Conference on Neural Networks, Washington, DC, USA, .
Williams, Ronald J. ; Peng, Jing. / Reinforcement learning algorithms as function optimizers. Paper presented at IJCNN International Joint Conference on Neural Networks, Washington, DC, USA, .7 p.
@conference{5b5ce1ba9df741f28969dc6a35f6772d,
title = "Reinforcement learning algorithms as function optimizers",
abstract = "Any nonassociative reinforcement learning algorithm can be viewed as a method for performing function optimization through (possibly noise-corrupted) sampling of function values. A description is given of the results of simulations in which the optima of several deterministic functions studied by D. H. Ackley (Ph.D. Diss., Carnegie-Mellon Univ., 1987) were sought using variants of REINFORCE algorithms. Results obtained for certain of these algorithms compare favorably to the best results found by Ackley.",
author = "Williams, {Ronald J.} and Jing Peng",
year = "1989",
month = "12",
day = "1",
language = "English",
pages = "89--95",
note = "null ; Conference date: 18-06-1989 Through 22-06-1989",

}

Williams, RJ & Peng, J 1989, 'Reinforcement learning algorithms as function optimizers' Paper presented at IJCNN International Joint Conference on Neural Networks, Washington, DC, USA, 18/06/89 - 22/06/89, pp. 89-95.

Reinforcement learning algorithms as function optimizers. / Williams, Ronald J.; Peng, Jing.

1989. 89-95 Paper presented at IJCNN International Joint Conference on Neural Networks, Washington, DC, USA, .

Research output: Contribution to conferencePaper

TY - CONF

T1 - Reinforcement learning algorithms as function optimizers

AU - Williams, Ronald J.

AU - Peng, Jing

PY - 1989/12/1

Y1 - 1989/12/1

N2 - Any nonassociative reinforcement learning algorithm can be viewed as a method for performing function optimization through (possibly noise-corrupted) sampling of function values. A description is given of the results of simulations in which the optima of several deterministic functions studied by D. H. Ackley (Ph.D. Diss., Carnegie-Mellon Univ., 1987) were sought using variants of REINFORCE algorithms. Results obtained for certain of these algorithms compare favorably to the best results found by Ackley.

AB - Any nonassociative reinforcement learning algorithm can be viewed as a method for performing function optimization through (possibly noise-corrupted) sampling of function values. A description is given of the results of simulations in which the optima of several deterministic functions studied by D. H. Ackley (Ph.D. Diss., Carnegie-Mellon Univ., 1987) were sought using variants of REINFORCE algorithms. Results obtained for certain of these algorithms compare favorably to the best results found by Ackley.

UR - http://www.scopus.com/inward/record.url?scp=0024874697&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0024874697

SP - 89

EP - 95

ER -

Williams RJ, Peng J. Reinforcement learning algorithms as function optimizers. 1989. Paper presented at IJCNN International Joint Conference on Neural Networks, Washington, DC, USA, .