This paper: 1) reviews different combinations between ANN's and evolutionary algorithms (EA's), including using EA's to evolve ANN connection weights, architectures, learning rules, and input features; 2) discusses different search operators which have been used in various EA's; and 3) points out possible future research directions. It is shown, through a considerably large literature review, that combinations between ANN's and EA's can lead to significantly better intelligent systems than relying on ANN's or EA's alone

Abstract | In this paper we describe the evolution of a discrete-time recurrent neural network to control a real mobile robot. In all our experiments the evolutionary procedure is carried out entirely on the physical robot without human intervention. We showthat the autonomous development of a set of behaviors for locating a battery charger and periodically returning to it can be achieved by lifting constraints in the design of the robot/environment interactions that were employed in a preliminary experiment. The emergent homing behavior is based on the autonomous development ofaninternal neural topographic map (which is not pre-designed) that allows the robot to choose the appropriate trajectory as function of location and remaining energy.

Research on potential interactions between connectionist learning systems, i.e., artificial neural networks (ANNs), and evolutionary search procedures, like genetic algorithms (GAs), has attracted a lot of attention recently. Evolutionary ANNs (EANNs) can be considered as the combination of ANNs and evolutionary search procedures. This paper first distinguishes among three kinds of evolution in EANNs, i.e., the evolution of connection weights, of architectures and of learning rules. Then it reviews each kind of evolution in detail and analyses critical issues related to different evolutions. The review shows that although a lot of work has been done on the evolution of connection weights and of architectures, few attempts have been made to understand the evolution of learning rules. Interactions among different evolutions are seldom mentioned in current research. However, the evolution of learning rules and its interactions with other kinds of evolution play a vital role in EANNs. As t...

Forma analysis is applied to the task of optimising the connectivity of a feed-forward neural network with a single layer of hidden units. This problem is reformulated as a multiset optimisation problem and techniques are developed to allow principled genetic search over fixed- and variable-size sets and multisets. These techniques require a further generalisation of the notion of gene, which is presented. The result is a non-redundant representation of the neural network topology optimisation problem together with recombination operators which have carefully designed and well-understood properties. The techniques developed have relevance to the application of genetic algorithms to constrained optimisation problems.

In this paper we present a new approach for automatic topology optimization of backpropagation networks. It is based on a genetic algorithm. In contrast to other approaches it allows that two networks with different number of units can be crossed to a new valid "child" network. We applied this algorithm to a medical classification task, which is extremely difficult to solve. The results confirm, that optimization make sence, because the generated network outperform all fixed topologies. 1 Introduction As Minsky and Papert [1] have shown the XOR-- problem cannot be solved without a hidden layer. A learning rule which is able to train this kind of networks was developed by Rumelhart et al. [2]. It is known as backpropagation (BP) and it is one of the most often used neural network paradigms. Hornik [3] proved that every function can be approximated by a neural network with just one hidden layer. By adding "enough" hidden units the approximation error can be made as small as required. Th...

This paper presents an evolutionary approach for the design of feed-forward and recurrent neural networks. We show that Evolutionary Algorithms can be used for the construction of networks for real world tasks. Therefore, a data-structure-based genotypic network representation, as well as genetic operators, are introduced. Results from the classification, function approximation and time-series domain are presented. 1. Introduction The performance of neural networks highly depends on the architecture of the networks and their parameters. Therefore, determeing the architecture of a network (size, structure, connectivity) greatly affects the performance criteria, i.e. learning speed, accuracy of learning, noise resistance, stability of fixed points and generalization ability. Recent works of Judd and Lin/Vitter show that learning in general, as well as choosing an optimal network topology, are NP-complete problems [8, 9]. They also have shown that placing constraints on the topology can...

this article we assume that the reader is familiar with the basic ideas of neural networks but perhaps less conversant with genetic algorithms. The aim is to describe the basic ideas of GAs and then to survey their application to the design of neural networks

Capturing higher-order principles of evolution and neural function in a model may result in simulation of the evolution of the brain and the associated behaviors it can produce. Accordingly, this work promotes the study of the evolution of artificial neural networks in virtual environments. Presented herein are arguments for the viability of such research, a model with the associated design and implementation of its simulator, and proposals for future pursuit. 1 INTRODUCTION Natural evolution produces a wide variety of organisms appropriately suited for survival in diverse ecological scenarios. In the evolutionary process, a structure eventually emerged that provided organisms a method of rapid internal communication and control---the neuron. Networks of these structures proved successful in coordinating environmental stimuli with an organism's actions, ultimately resulting in the brain. Ethologists study animal behavior in natural environments, stressing adaptation and evolution. Neu...

this paper is to provide a brief overview of some of our projects involving the application of evolutionary algorithms to industrial application problems. These projects are established at the Center for Applied Systems Analysis (Casa) within the Informatik Centrum Dortmund (ICD), a research institute providing technology transfer from university research into industrial practice. Presently, 1