Adaptive stochastic cellular automata: Theory

Y. C. Lee; S. Qian; R. D. Jones; C. W. Barnes; G. W. Flake; M. K. O'Rourke; K. Lee; H. H. Chen; G. Z. Sun; Y. Q. Zhang; D. Chen; C. L. Giles

doi:10.1016/0167-2789(90)90180-W

Adaptive stochastic cellular automata: Theory

Y. C. Lee, S. Qian, R. D. Jones, C. W. Barnes, G. W. Flake, M. K. O'Rourke, K. Lee, H. H. Chen, G. Z. Sun, Y. Q. Zhang, D. Chen, C. L. Giles

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

15 Scopus citations

Abstract

The mathematical concept of cellular automata has been generalized to allow for the possibility that the uniform local interaction rules that govern conventional cellular automata are replaced by nonuniform local interaction rules which are drawn from the same probability distribution function, in order to guarantee the statistical homogeneity of the cellular automata system. Adaptation and learning in such a system can be accomplished by evolving the probability distribution function along the steepest descent direction of some objective function in a statistically unbiased way to ensure that the cellular automata's dynamical behavior approaches the desired behavior asymptotically. The proposed CA model has been shown mathematically to possess the requisite convergence property under general conditions.

Original language	English (US)
Pages (from-to)	159-180
Number of pages	22
Journal	Physica D: Nonlinear Phenomena
Volume	45
Issue number	1-3
DOIs	https://doi.org/10.1016/0167-2789(90)90180-W
State	Published - Sep 2 1990

All Science Journal Classification (ASJC) codes

Statistical and Nonlinear Physics
Mathematical Physics
Condensed Matter Physics
Applied Mathematics

Access to Document

10.1016/0167-2789(90)90180-W

Cite this

@article{b7120d3decc941ebbbecb6a053abb1e3,

title = "Adaptive stochastic cellular automata: Theory",

abstract = "The mathematical concept of cellular automata has been generalized to allow for the possibility that the uniform local interaction rules that govern conventional cellular automata are replaced by nonuniform local interaction rules which are drawn from the same probability distribution function, in order to guarantee the statistical homogeneity of the cellular automata system. Adaptation and learning in such a system can be accomplished by evolving the probability distribution function along the steepest descent direction of some objective function in a statistically unbiased way to ensure that the cellular automata's dynamical behavior approaches the desired behavior asymptotically. The proposed CA model has been shown mathematically to possess the requisite convergence property under general conditions.",

author = "Lee, {Y. C.} and S. Qian and Jones, {R. D.} and Barnes, {C. W.} and Flake, {G. W.} and O'Rourke, {M. K.} and K. Lee and Chen, {H. H.} and Sun, {G. Z.} and Zhang, {Y. Q.} and D. Chen and Giles, {C. L.}",

year = "1990",

month = sep,

day = "2",

doi = "10.1016/0167-2789(90)90180-W",

language = "English (US)",

volume = "45",

pages = "159--180",

journal = "Physica D: Nonlinear Phenomena",

issn = "0167-2789",

publisher = "Elsevier",

number = "1-3",

}

TY - JOUR

T1 - Adaptive stochastic cellular automata

T2 - Theory

AU - Lee, Y. C.

AU - Qian, S.

AU - Jones, R. D.

AU - Barnes, C. W.

AU - Flake, G. W.

AU - O'Rourke, M. K.

AU - Lee, K.

AU - Chen, H. H.

AU - Sun, G. Z.

AU - Zhang, Y. Q.

AU - Chen, D.

AU - Giles, C. L.

PY - 1990/9/2

Y1 - 1990/9/2

N2 - The mathematical concept of cellular automata has been generalized to allow for the possibility that the uniform local interaction rules that govern conventional cellular automata are replaced by nonuniform local interaction rules which are drawn from the same probability distribution function, in order to guarantee the statistical homogeneity of the cellular automata system. Adaptation and learning in such a system can be accomplished by evolving the probability distribution function along the steepest descent direction of some objective function in a statistically unbiased way to ensure that the cellular automata's dynamical behavior approaches the desired behavior asymptotically. The proposed CA model has been shown mathematically to possess the requisite convergence property under general conditions.

AB - The mathematical concept of cellular automata has been generalized to allow for the possibility that the uniform local interaction rules that govern conventional cellular automata are replaced by nonuniform local interaction rules which are drawn from the same probability distribution function, in order to guarantee the statistical homogeneity of the cellular automata system. Adaptation and learning in such a system can be accomplished by evolving the probability distribution function along the steepest descent direction of some objective function in a statistically unbiased way to ensure that the cellular automata's dynamical behavior approaches the desired behavior asymptotically. The proposed CA model has been shown mathematically to possess the requisite convergence property under general conditions.

UR - http://www.scopus.com/inward/record.url?scp=0042731528&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0042731528&partnerID=8YFLogxK

U2 - 10.1016/0167-2789(90)90180-W

DO - 10.1016/0167-2789(90)90180-W

M3 - Article

AN - SCOPUS:0042731528

SN - 0167-2789

VL - 45

SP - 159

EP - 180

JO - Physica D: Nonlinear Phenomena

JF - Physica D: Nonlinear Phenomena

IS - 1-3

ER -

Adaptive stochastic cellular automata: Theory

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this