Artificial neural networks activation function

These considerations provide an impetus for the development of fast, nonlinear, variable selection QSAR methods that can avoid the aforementioned problems of linear QSAR. Several nonlinear QSAR methods have been proposed in recent years. Most of these methods are based on either artificial neural network (ANN) (50, 61, 137-142) or machine learning techniques (65,143-145). Given that optimization of many parameters is involved in these techniques, the speed of the analysis is relatively slow. More recently. Hirst reported a simple and fast nonlinear QSAR method (146), in which the activity surface was generated from the activities of training set compounds based on some predefined mathematical function. 

Fig. 4.19 Artificial neural network (ANN) and SUBSTRUCT classification of CNS activity. Result of the ANN classification of CNS activity (left) the correct classification rate is given as a function of the limit separating CNS-i- and CNS- molecules, respectively. For example, if the limit is 0.5 all molecules with a score <0.5 are classified as...

It is well known that electrical conductivity of liquid solutions depends on the concentration of ions and their activity. The aqueous fluids in pipelines usually are electrolyte solutions and the conductivity is proportional to the salt concentration. The activity of the ions is related to temperature, and impurity like nonconductive chemical additives. Measurements of electrical conductivity could directly reflect the concentrations of chemicals such as salts, THIs (alcohol) and KHIs (polymers). (Clay and Medwin, 1977) [6] presented a simple correlation in which the sound velocity in sea water was described as a function of sahnity and temperature. Acoustic velocity has been successfully applied to investigate a variety of solutions and binary gas mixtures (Jerie, et al., 2004 Vibhu, et al., 2004 Goodenough, et al., 2005 Vyas, et al., 2006) [11] [25] [9] [26]. As a result, electrical conductivity and acoustic velocity were chosen as two parameters to simultaneously determine both salt and inhibitor concentrations. Artificial neural network (ANN) provides a numerical tool for such applications in which multi-parameter correlations are needed but the interaction and the relations between the parameters are not well known (Sundgren, et al., 1991 Broten and Wood, 1993) [21] [4]. Therefore, ANN correlations were developed to determine salt and inhibitor concentrations using the measured electrical conductivity, acoustic velocity, and temperature. 

A multilayer perceptron (MLP) is a feed-forward artificial neural network model that maps sets of input data onto a set of suitable outputs (Patterson 1998). A MLP consists of multiple layers of nodes in a directed graph, with each layer fully connected to the next one. Except for the input nodes, each node is a neuron (or processing element) with a nonlinear activation function. MLP employs a supervised learning techruque called backpropagation for training the network. MLP is a modification of the standard linear perceptron and can differentiate data that are not linearly separable. 

Also in chemistry artificial neural networks have found wide use. They have been used to fit spectroscopic data, to investigate quantitative structure-activity relationships (QSAR), to predict deposition rates in chemical vapor deposition, to predict binding sites of biomolecules, to derive pair potentials from diffraction data on liquids, " to solve the Schrodinger equation for simple model potentials like the harmonic oscillator, to estimate the fitness function in genetic algorithm optimizations, in experimental data analysis, to predict the secondary structure of proteins, to predict atomic energy levels, " and to solve classification problems from clinical chemistry, in particular the differentiation between diseases on the basis of characteristic laboratory data. ... 

The multilayer perceptron is thus an artificial neural network consisting of M layers. Each neuron is a simple processing unit it performs weighted summation of the inputs, calculates the excitation and generates the value specified by the activation function and the stimulation value in the output channel ... 

---