Neural network examples

Artificial Neural Networks. An Artificial Neural Network (ANN) consists of a network of nodes (processing elements) connected via adjustable weights [Zurada, 1992]. The weights can be adjusted so that a network learns a mapping represented by a set of example input/output pairs. An ANN can in theory reproduce any continuous function 95 —>31 °, where n and m are numbers of input and output nodes. In NDT neural networks are usually used as classifiers... 

Hybrid systems. Depending on the problem to be solved, use can also be made of a combination of techniques leading to a hybrid system. For example, a rule-based system may use neural networks for solving classification subproblems (as is described in [Hopgood, 1993]), or a combination of a rule-based and a CBR system can be used as in the system for URS data interpretation described later in this paper. 

Neural network classifiers. The neural network or other statistical classifiers impose strong requirements on the data and the inspection, however, when these are fulfilled then good fully automatic classification systems can be developed within a short period of time. This is for example the case if the inspection is a part of a manufacturing process, where the inspected pieces and the possible defect mechanisms are well known and the whole NDT inspection is done in repeatable conditions. In such cases it is possible to collect (or manufacture) as set of defect pieces, which can be used to obtain a training set. There are some commercially available tools (like ICEPAK [Chan, et al., 1988]) which can construct classifiers without any a-priori information, based only on the training sets of data. One has, however, always to remember about the limitations of this technique, otherwise serious misclassifications may go unnoticed. 

Let us start with a classic example. We had a dataset of 31 steroids. The spatial autocorrelation vector (more about autocorrelation vectors can be found in Chapter 8) stood as the set of molecular descriptors. The task was to model the Corticosteroid Ringing Globulin (CBG) affinity of the steroids. A feed-forward multilayer neural network trained with the back-propagation learning rule was employed as the learning method. The dataset itself was available in electronic form. More details can be found in Ref. [2]. 

Chirality codes are used to represent molecular chirality by a fixed number of de-.scriptors. Thc.se descriptors can then be correlated with molecular properties by way of statistical methods or artificial neural networks, for example. The importance of using descriptors that take different values for opposite enantiomers resides in the fact that observable properties are often different for opposite enantiomers. 

As explained in Chapter 8, descriptors are used to represent a chemical structure and, thus, to provide a coding which allows electronic processing of chemical data. The example given here shows how a GA is used to Rnd an optimal set of descriptors for the task of classification using a Kohoncii neural network. The chromosomes of the GA are to be used as a means for selecting the descriptors they indicate which descriptors are used and which are rejected ... 

Neural networks model the functionality of the brain. They learn from examples, whereby the weights of the neurons are adapted on the basis of training data. 

Neural networks can learn automatically from a data set of examples. In the case of NMR chemical shiffs, neural networks have been trained to predict the chemical shift of protons on submission of a chemical structure. Two main issues play decisive roles how a proton is represented, and which examples are in the data set. 

An example of the neural network prediction of NMR chemical shifts for a natural product is illustrated in Figure 10.2-7 together with the calculations from other methods. This molecule was chosen as it had been discovered [47]... 

The results presented here imply that a similar approadi can be used for comparing two different hbraries, for determining the degree of overlap between the compounds in these two Hbraries. Examples of the application of artificial neural networks or GA in drug design are given in [57, 58, 84, 85]. 

VR, the inputs correspond to the value of the various parameters and the network is 1 to reproduce the experimentally determined activities. Once trained, the activity of mown compound can be predicted by presenting the network with the relevant eter values. Some encouraging results have been reported using neural networks, have also been applied to a wide range of problems such as predicting the secondary ire of proteins and interpreting NMR spectra. One of their main advantages is an to incorporate non-linearity into the model. However, they do present some problems Hack et al. 1994] for example, if there are too few data values then the network may memorise the data and have no predictive capability. Moreover, it is difficult to the importance of the individual terms, and the networks can require a considerable 1 train. 

Neural networks have the following advantages (/) once trained, their response to input data is extremely fast (2) they are tolerant of noisy and incomplete input data (J) they do not require knowledge engineering and can be built direcdy from example data (4) they do not require either domain models or models of problem solving and (5) they can store large amounts of information implicitly. 

The first is the relational model. Examples are hnear (i.e., models linear in the parameters and neural network models). The model output is related to the input and specifications using empirical relations bearing no physical relation to the actual chemical process. These models give trends in the output as the input and specifications change. Actual unit performance and model predictions may not be very close. Relational models are usebil as interpolating tools. 

Intended Use The intended use of the model sets the sophistication required. Relational models are adequate for control within narrow bands of setpoints. Physical models are reqiiired for fault detection and design. Even when relational models are used, they are frequently developed bv repeated simulations using physical models. Further, artificial neural-network models used in analysis of plant performance including gross error detection are in their infancy. Readers are referred to the work of Himmelblau for these developments. [For example, see Terry and Himmelblau (1993) cited in the reference list.] Process simulators are in wide use and readily available to engineers. Consequently, the emphasis of this section is to develop a pre-liminaiy physical model representing the unit. 

Fig. 10.43 Neural network structure for Example 10.9. Hidden layer...

This tutorial uses the MATLAB Control System Toolbox, the Fuzzy Logie Toolbox and the Neural Network Toolbox. Problems in Chapter 10 are used as design examples. 

The MATLAB Neural Network Toolbox This Toolbox was not used in the Examples given in Chapter 10. For details on the Toolbox, type... 

Even so, artificial neural networks exhibit many brainlike characteristics. For example, during training, neural networks may construct an internal mapping/ model of an external system. Thus, they are assumed to make sense of the problems that they are presented. As with any construction of a robust internal model, the external system presented to the network must contain meaningful information. In general the following anthropomorphic perspectives can be maintained while preparing the data ... 

The nearest neighbour method is often applied, with, in view of its simplicity, surprisingly good results. An example where -NN performs well in a comparison with neural networks and SIMCA (see further) can be found in [16]. 

A more recently introduced technique, at least in the field of chemometrics, is the use of neural networks. The methodology will be described in detail in Chapter 44. In this chapter, we will only give a short and very introductory description to be able to contrast the technique with the others described earlier. A typical artificial neuron is shown in Fig. 33.19. The isolated neuron of this figure performs a two-stage process to transform a set of inputs in a response or output. In a pattern recognition context, these inputs would be the values for the variables (in this example, limited to only 2, X and x- and the response would be a class variable, for instance y = 1 for class K and y = 0 for class L. 

---