Neural Network/Evolutionary Algorithm

In general, simulation methods can be divided into those that use a computer and those that do not, as shown in Fig. 14.1. Those simulations without computer can be separated into destructive and nondestructive methods. Simulations that use a computer can either be based on technical models (for example, finite element method for structural composites, computational fluid dynamics for polymer flows), on examples taken from nature (for example, artificial neural networks, evolutionary algorithms for machine setting optimization), and those based on analytical equations (for example, warp tension during weaving). 

Beside trial and error methods, other methodologies were tested in order to obtain the best performance of the neural models evolutionary algorithms represent appropriate methods for determining optimal neural network structure. 

Concomitantly with the increase in hardware capabilities, better software techniques will have to be developed. It will pay us to continue to learn how nature tackles problems. Artificial neural networks are a far cry away from the capabilities of the human brain. There is a lot of room left from the information processing of the human brain in order to develop more powerful artificial neural networks. Nature has developed over millions of years efficient optimization methods for adapting to changes in the environment. The development of evolutionary and genetic algorithms will continue. 

B. Carse and T.C. Fogarty, Fast evolutionary learning of minimal radial basis function neural networks using a genetic algorithm. Lecture Notes in Computer Science, 1143, (1996) 1-22. 

Just as there are several varieties of evolutionary algorithm, so the neural network is available in several flavors. We shall consider feedforward networks and, briefly, Kohonen networks and growing cell structures, but Hop-field networks, which we shall not cover in this chapter, also find some application in science.31... 

The four-volume Handbook of Chemoinformatics—From Data to Knowledge (Gasteiger 2003) contains a number of introductions and reviews that are relevant to chemometrics Partial Least Squares (PLS) in Cheminformatics (Eriksson et al. 2003), Inductive Learning Methods (Rose 1998), Evolutionary Algorithms and their Applications (von Homeyer 2003), Multivariate Data Analysis in Chemistry (Varmuza 2003), and Neural Networks (Zupan 2003). 

The boundaries to AI, within which evolutionary optimisers He, are woolly, and this may make it appear a mysterious subject to those who are unfamiliar with it even those whose research lies in AI find it hard to agree on the precise boundaries to the field. It encompasses a large family of algorithms, from neural networks and knowledge-based systems to those within evolutionary computing. Almost every major AI technique is now used within chemistry (Table 1), and in an increasing number of cases, AI is the method of choice. 

The neural network architecture optimized by the evolutionary algorithm could be analyzed for a biochemical interpretation and feature extraction. One may infer the importance of input properties based on the relative connectivity of the input units. For example, bulkiness, which was not connected at all, was probably unimportant. On the other hand, units for polarity, refractivity, hydrophobicity, and surface area were highly connected, indicating these are important features of membrane transition regions. 

Sundaram, A. Ghosh, P. Caruthers, J. Venkatasubramanian, V. Design of fuel additives using neural networks and evolutionary algorithms. AIChE J. 2001, 47, 1387-1406. 

Farina, M. (2002). A neural network based generalized response surface multiobjective evolutionary algorithm, in Proceedings of IEEE World Congress on Computational Intelligence (WCCI-2002) (Hawaii). 

As with all other evolutionary algorithms, the initial population is generated at random. The exact manner in which this happens, however, is entirely dependent on the problem domain. EP can handle any structure, whether as simple as the binary strings used by GAs, sets of real numbers, or even highly complex structures like neural networks. 

---