Conference paper

Adaptation of Neural Networks Using Genetic Algorithms

Tin Ilakovac ; Ruđer Bošković Institute, POB 1016, HR-41001 Zagreb, Croatia

Abstract

This paper proposes an approach to the problem of adaptation of neuralnetworks (NN) to arbitrary tasks. It is well known that the functionalproperties of a NN depend on its construction: on topological structure,learning and activation methods, and signal output. A definition language is developed for describing various constructions of NNs in the shape of strings. This paper uses a model of a neuron which has a receptive field and adaptable learning, activation and signaling, while the NN model consists of interconnected layers allowing feedforward, feedback and lateral connections with a single input and ouput layer. Adaptation of NNs is done with a genetic algorithm (GA)using crossover, mutation, and proportional
selection operators on a population of strings that represent NNs. These strings (and their respective NNs) are evolved until they find solutions to given tasks which are defined as objective functions. The paper proposes a solution to »deception«,an important problem concerning GA's convergence: a strict hierarchy in the description of NNs based on ordered express ion which decreases the probability of dual representations. This approach can develop autodidactive NNs.

Keywords

Hrčak ID:

176542

URI

https://hrcak.srce.hr/176542

Publication date:

1.2.1995.

Visits: 757 *