Results 1 
5 of
5
Random Perturbations to Hebbian Synapses of Associative Memory using a Genetic Algorithm
, 1997
"... . We apply evolutionary algorithms to Hopfield model of associative memory. Previously we reported that a genetic algorithm using ternary chromosomes evolves the Hebbrule associative memory to enhance its storage capacity by pruning some connections. This paper describes a genetic algorithm using ..."
Abstract

Cited by 4 (4 self)
 Add to MetaCart
. We apply evolutionary algorithms to Hopfield model of associative memory. Previously we reported that a genetic algorithm using ternary chromosomes evolves the Hebbrule associative memory to enhance its storage capacity by pruning some connections. This paper describes a genetic algorithm using realencoded chromosomes which successfully evolves overloaded Hebbian synaptic weights to function as an associative memory. The goal of this study is to shed new light on the analysis of the Hopfield model, which also enables us to use the model as more challenging test suite for evolutionary computations. 1 Introduction In the field of evolutionary computations, a lot of complicated functions have been proposed as test suites for comparing and evaluating the effectiveness of various evolutionary computations [1, 2, 3, 4, 5, 6, 7, 8]. However as Mulenbein [9] pointed out: But are such problems typical applications? We have not encountered such a problem. these functions are far from re...
Evolution of a Hopfield Associative Memory by the Breeder Genetic Algorithm
 Proceedings of the Seventh International Conference on Genetic Algorithms
, 1997
"... We apply some variants of evolutionary computations to the Hopfield model of associative memory. In this paper, we use the Breeder Genetic Algorithm (BGA) to explore the optimal set of synaptic weights with respect to the storage capacity. We present the BGA has tremendous ability to search a soluti ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
We apply some variants of evolutionary computations to the Hopfield model of associative memory. In this paper, we use the Breeder Genetic Algorithm (BGA) to explore the optimal set of synaptic weights with respect to the storage capacity. We present the BGA has tremendous ability to search a solution in the massively multimodal landscape of the synaptic weight space. The main goal of this study is to shed new light on the analysis of the Hopfield model of associative memory. We also expect the model to be used as a new test function of evolutionary computations. 1 INTRODUCTION Associative memory is a dynamical system which has a number of stable states with a domain of attraction around them [1]. If the system starts at any state in the domain, it will converge to the stable state. In 1982, Hopfield [2] proposed a fully connected neural network model of associative memory in which we can store information by distributing it among neurons, and we can recall it from its noisy and/or p...
How Realvalued Random Synapses Evolve toward Symmetry with Diploid Chromosomes?
 Diploid Chromosomes? International ICSC/IFAC Symposium on Neural Computation
, 1998
"... We apply genetic algorithms to the Hopfield's neural network model of associative memory. Previously, we observed that random synaptic weights of a network evolved to create the fixed point attractors in pattern space exactly at the locations corresponding to a set of given patterns to be stored. In ..."
Abstract

Cited by 1 (1 self)
 Add to MetaCart
We apply genetic algorithms to the Hopfield's neural network model of associative memory. Previously, we observed that random synaptic weights of a network evolved to create the fixed point attractors in pattern space exactly at the locations corresponding to a set of given patterns to be stored. In that simulation, a weight configuration was represented as a sequence of genes which might be called a haploid chromosome. Then, a population of these haploid chromosomes underwent evolution. In this paper, we employ diploid chromosomes, a pair of chromosomes. We have reported elsewhere two different versions of this evolution. One is the evolution of the Hebbian synapses in which some of the synapses are adaptively pruned according to information in diploid chromosomes. The other is the evolution of random synaptic weights whose values are restricted to 61, and are encoded directly into diploid chromosomes. Here, we describe the evolution of realvalued random synaptic weights. keywords ...
Does Diploidy Affect Evolutions of Hopfield Associative Memory?
"... . We apply genetic algorithms to the Hopfield's neural network model of associative memory. Previously, we observed that random synaptic weights of a network evolved to create the fixed point attractors corresponding to a set of given patterns to be stored. In that simulation, a weight configuration ..."
Abstract

Cited by 1 (1 self)
 Add to MetaCart
. We apply genetic algorithms to the Hopfield's neural network model of associative memory. Previously, we observed that random synaptic weights of a network evolved to create the fixed point attractors corresponding to a set of given patterns to be stored. In that simulation, a weight configuration was expressed as a sequence of genes which might be called a haploid chromosome. Then a population of these haploid chromosomes underwent evolution. In this paper, we employ diploid chromosomes, a pair of chromosomes. We have reported elsewhere two different versions of the evolution that uses diploid chromosomes. One is the evolution of the Hebbian synapses in which some of the synapses are adaptively pruned according to information in the diploid chromosomes. The other is the evolution of realvalued random synaptic weights which are encoded directly into diploid chromosomes. Here, we describe the evolution based on the latter scheme with the weights being restricted to take the value 61....
Evolution of Associative Memory using Diploid Chromosomes
, 1997
"... . We apply genetic algorithms to the Hopfield's neural network model of associative memory. Previously, we reported that random synaptic weights of the network evolved to create the fixed point attractors exactly at the locations of given patterns. Furthermore, we reported that the genetic algorithm ..."
Abstract
 Add to MetaCart
. We apply genetic algorithms to the Hopfield's neural network model of associative memory. Previously, we reported that random synaptic weights of the network evolved to create the fixed point attractors exactly at the locations of given patterns. Furthermore, we reported that the genetic algorithm can evolve the Hebbian synaptic weights to enlarge the storage capacity. In those experiments, the genetic algorithm pruned a certain fraction of the synaptic connections adaptively with using haploid chromosomes. In this paper, we present the evolution can also be made by using diploid chromosomes. 1 Introduction Associative memory is a dynamical system which has a number of stable states with a domain of attraction around them [1]. If the system starts at any state in the domain, it will converge to the stable state. In 1982, Hopfield [2] proposed a fully connected neural network model of associative memory in which we can store information by distributing it among neurons, and we can re...