Using graph theoretic measures to predict the performance of associative memory models

L. Calcraft; R.G. Adams; W. Chen; N. Davey

Using graph theoretic measures to predict the performance of associative memory models

L. Calcraft, R.G. Adams, W. Chen, N. Davey

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

33 Downloads (Pure)

Abstract

We test a selection of associative memory models built with different connection strategies, exploring the relationship between the structural properties of each network and its pattern-completion performance. It is found that the Local Efficiency of the network can be used to predict pattern completion performance for associative memory models built with a range of different connection strategies. This relationship is maintained as the networks are scaled up in size, but breaks down under conditions of very sparse connectivity.

Original language	English
Title of host publication	ESANN2008: 16th European Symposium on Artificial Neural Networks
Publisher	ESANN
Pages	107-112
ISBN (Print)	2-930-307080
Publication status	Published - 2008

Access to Document

902639.pdf

Cite this

@inproceedings{ed66b94cc6384452926f4eba5bb00e07,

title = "Using graph theoretic measures to predict the performance of associative memory models",

abstract = "We test a selection of associative memory models built with different connection strategies, exploring the relationship between the structural properties of each network and its pattern-completion performance. It is found that the Local Efficiency of the network can be used to predict pattern completion performance for associative memory models built with a range of different connection strategies. This relationship is maintained as the networks are scaled up in size, but breaks down under conditions of very sparse connectivity.",

author = "L. Calcraft and R.G. Adams and W. Chen and N. Davey",

note = "Original paper can be found at: http://www.dice.ucl.ac.be/esann/proceedings/electronicproceedings.htm",

year = "2008",

language = "English",

isbn = "2-930-307080",

pages = "107--112",

booktitle = "ESANN2008: 16th European Symposium on Artificial Neural Networks",

publisher = "ESANN",

}

TY - GEN

T1 - Using graph theoretic measures to predict the performance of associative memory models

AU - Calcraft, L.

AU - Adams, R.G.

AU - Chen, W.

AU - Davey, N.

N1 - Original paper can be found at: http://www.dice.ucl.ac.be/esann/proceedings/electronicproceedings.htm

PY - 2008

Y1 - 2008

N2 - We test a selection of associative memory models built with different connection strategies, exploring the relationship between the structural properties of each network and its pattern-completion performance. It is found that the Local Efficiency of the network can be used to predict pattern completion performance for associative memory models built with a range of different connection strategies. This relationship is maintained as the networks are scaled up in size, but breaks down under conditions of very sparse connectivity.

AB - We test a selection of associative memory models built with different connection strategies, exploring the relationship between the structural properties of each network and its pattern-completion performance. It is found that the Local Efficiency of the network can be used to predict pattern completion performance for associative memory models built with a range of different connection strategies. This relationship is maintained as the networks are scaled up in size, but breaks down under conditions of very sparse connectivity.

M3 - Conference contribution

SN - 2-930-307080

SP - 107

EP - 112

BT - ESANN2008: 16th European Symposium on Artificial Neural Networks

PB - ESANN

ER -

Using graph theoretic measures to predict the performance of associative memory models

Abstract

Access to Document

Fingerprint

Cite this