A Rule Chaining Architecture Using a Correlation Matrix Memory

Jim Austin; Stephen John Hobson; Nathan John Burles; Simon O'Keefe

doi:10.1007/978-3-642-33269-2_7

A Rule Chaining Architecture Using a Correlation Matrix Memory

Jim Austin, Stephen John Hobson, Nathan John Burles, Simon O'Keefe

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Chapter (peer-reviewed) › peer-review

Abstract

This paper describes an architecture based on superimposed distributed representations and distributed associative memories which is capable of performing rule chaining. The use of a distributed representation allows the system to utilise memory efficiently, and the use of superposition reduces the time complexity of a tree search to O(d), where d is the depth of the tree. Our experimental results show that the architecture is capable of rule chaining effectively, but that further investigation is needed to address capacity considerations.

Original language	English
Title of host publication	Artificial Neural Networks and Machine Learning – ICANN 2012
Subtitle of host publication	22nd International Conference on Artificial Neural Networks, Lausanne, Switzerland, September 11-14, 2012, Proceedings, Part I
Publisher	Springer
Pages	49-56
Volume	7552
Edition	PART 1
ISBN (Electronic)	978-3-642-33269-2
ISBN (Print)	978-3-642-33268-5
DOIs	https://doi.org/10.1007/978-3-642-33269-2_7
Publication status	Published - 2012
Event	22nd International Conference on Artificial Neural Networks, ICANN 2012 - Lausanne, Switzerland Duration: 11 Sept 2012 → 14 Sept 2012

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	7552
ISSN (Print)	0302-9743

Conference

Conference	22nd International Conference on Artificial Neural Networks, ICANN 2012
Country/Territory	Switzerland
City	Lausanne
Period	11/09/12 → 14/09/12

Access to Document

10.1007/978-3-642-33269-2_7

ruleChainingSystem
© Springer-Verlag, 2012. This is an author-produced version of the published paper. Uploaded in accordance with the publisher’s self-archiving policy. Further copying may not be permitted; contact the publisher for details
Submitted manuscript, 291 KB

Cite this

Austin, J., Hobson, S. J., Burles, N. J., & O'Keefe, S. (2012). A Rule Chaining Architecture Using a Correlation Matrix Memory. In Artificial Neural Networks and Machine Learning – ICANN 2012: 22nd International Conference on Artificial Neural Networks, Lausanne, Switzerland, September 11-14, 2012, Proceedings, Part I (PART 1 ed., Vol. 7552, pp. 49-56). (Lecture Notes in Computer Science; Vol. 7552). Springer. https://doi.org/10.1007/978-3-642-33269-2_7

Austin, Jim ; Hobson, Stephen John ; Burles, Nathan John et al. / A Rule Chaining Architecture Using a Correlation Matrix Memory. Artificial Neural Networks and Machine Learning – ICANN 2012: 22nd International Conference on Artificial Neural Networks, Lausanne, Switzerland, September 11-14, 2012, Proceedings, Part I. Vol. 7552 PART 1. ed. Springer, 2012. pp. 49-56 (Lecture Notes in Computer Science).

@inbook{3ecc43e3eca74087ba2a6c8d3a03b11f,

title = "A Rule Chaining Architecture Using a Correlation Matrix Memory",

abstract = "This paper describes an architecture based on superimposed distributed representations and distributed associative memories which is capable of performing rule chaining. The use of a distributed representation allows the system to utilise memory efficiently, and the use of superposition reduces the time complexity of a tree search to O(d), where d is the depth of the tree. Our experimental results show that the architecture is capable of rule chaining effectively, but that further investigation is needed to address capacity considerations.",

author = "Jim Austin and Hobson, {Stephen John} and Burles, {Nathan John} and Simon O'Keefe",

year = "2012",

doi = "10.1007/978-3-642-33269-2_7",

language = "English",

isbn = "978-3-642-33268-5",

volume = "7552",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "49--56",

booktitle = "Artificial Neural Networks and Machine Learning – ICANN 2012",

address = "Germany",

edition = "PART 1",

note = "22nd International Conference on Artificial Neural Networks, ICANN 2012 ; Conference date: 11-09-2012 Through 14-09-2012",

}

Austin, J , Hobson, SJ , Burles, NJ & O'Keefe, S 2012, A Rule Chaining Architecture Using a Correlation Matrix Memory. in Artificial Neural Networks and Machine Learning – ICANN 2012: 22nd International Conference on Artificial Neural Networks, Lausanne, Switzerland, September 11-14, 2012, Proceedings, Part I. PART 1 edn, vol. 7552, Lecture Notes in Computer Science, vol. 7552, Springer, pp. 49-56, 22nd International Conference on Artificial Neural Networks, ICANN 2012, Lausanne, Switzerland, 11/09/12. https://doi.org/10.1007/978-3-642-33269-2_7

A Rule Chaining Architecture Using a Correlation Matrix Memory. / Austin, Jim ; Hobson, Stephen John ; Burles, Nathan John et al.
Artificial Neural Networks and Machine Learning – ICANN 2012: 22nd International Conference on Artificial Neural Networks, Lausanne, Switzerland, September 11-14, 2012, Proceedings, Part I. Vol. 7552 PART 1. ed. Springer, 2012. p. 49-56 (Lecture Notes in Computer Science; Vol. 7552).

Research output: Chapter in Book/Report/Conference proceeding › Chapter (peer-reviewed) › peer-review

TY - CHAP

T1 - A Rule Chaining Architecture Using a Correlation Matrix Memory

AU - Austin, Jim

AU - Hobson, Stephen John

AU - Burles, Nathan John

AU - O'Keefe, Simon

PY - 2012

Y1 - 2012

N2 - This paper describes an architecture based on superimposed distributed representations and distributed associative memories which is capable of performing rule chaining. The use of a distributed representation allows the system to utilise memory efficiently, and the use of superposition reduces the time complexity of a tree search to O(d), where d is the depth of the tree. Our experimental results show that the architecture is capable of rule chaining effectively, but that further investigation is needed to address capacity considerations.

AB - This paper describes an architecture based on superimposed distributed representations and distributed associative memories which is capable of performing rule chaining. The use of a distributed representation allows the system to utilise memory efficiently, and the use of superposition reduces the time complexity of a tree search to O(d), where d is the depth of the tree. Our experimental results show that the architecture is capable of rule chaining effectively, but that further investigation is needed to address capacity considerations.

UR - http://www.scopus.com/inward/record.url?scp=84867677483&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-33269-2_7

DO - 10.1007/978-3-642-33269-2_7

M3 - Chapter (peer-reviewed)

SN - 978-3-642-33268-5

VL - 7552

T3 - Lecture Notes in Computer Science

SP - 49

EP - 56

BT - Artificial Neural Networks and Machine Learning – ICANN 2012

PB - Springer

T2 - 22nd International Conference on Artificial Neural Networks, ICANN 2012

Y2 - 11 September 2012 through 14 September 2012

ER -

Austin J , Hobson SJ , Burles NJ , O'Keefe S. A Rule Chaining Architecture Using a Correlation Matrix Memory. In Artificial Neural Networks and Machine Learning – ICANN 2012: 22nd International Conference on Artificial Neural Networks, Lausanne, Switzerland, September 11-14, 2012, Proceedings, Part I. PART 1 ed. Vol. 7552. Springer. 2012. p. 49-56. (Lecture Notes in Computer Science). doi: 10.1007/978-3-642-33269-2_7

A Rule Chaining Architecture Using a Correlation Matrix Memory

Abstract

Publication series

Conference

Access to Document

Other files and links

Cite this