Artificial neural networks topology

Z. Roger, R. Weber, Finding an optimal artificial neural network topology in real-life modeling, presented at the ICSC Symposium on Neural Computation, article No. 1, 1403/109, 2000. 

Ivanciuc, O. (2000g) Molecular structure encoding into artificial neural networks topology. Roum. Chem. Quart. Rev, 8, 197-220. 

Bevilacqua, V., Mastronardi, G., Menolascina, F., Pannaxale, P., Pedone, A. A novel multi-objective genetic algorithm approach to artificial neural network topology optimisation the breast cancer classification probiem. In International Joint Conference on Neural Networks, pp. 1958-1965. IEEE (2006)... 

Muecia-Soler, M., Perez-Gimenez, F., GAECiA-MARCH, F.J., Salabert-Salvador, M.T., Diaz-Villanueva, W., and Castro-Bleda, M.J. Dmgs and nondmgs an effective discrimination with topological methods and artificial neural networks. /. Chem. Inf. Comput.. Sci. 2003, 43, 1688-1702. 

Figure6.25 Schematicdrawingofan artificial neural network with a multilayer perceptron topology, showing the pathways from the input Xj to the output y , and the visible and hidden node layers.

Description A QSAR toolkit with descriptor generation (topological, geometrical, electronic, and physicochemical descriptors), variable selection, regression and artificial neural network modelling. 

PCA can be used, for instance, to select a smaller set of descriptors to be used for multivariate regression and artificial neural networks. A set of descriptors can be divided into subclasses, such as geometrical, topological, or constitutional. A principal component analysis including the required output can then be performed for each subclass and the descriptor nearest to the output can be selected, checking if it improved the predictive power of a model. 

Fig. 4.14 Topology of a three-layered artificial neural network (ANN). The information flow is from left to right. The parameters representing a given compound are read into the input layer neurons from where the information is fed forward (weighted by v,j) to the neurons in the hidden layer, and so on. For clarity, not all connections...

FI C U RE 4 The topology of artificial neural network used in this study. 

Rose, V. S., Macfie, H. J. H., and Croall, I. F. (1991) Kohonen topology-preserving mapping an unsupervised artificial neural network method for use in QSAR analysis. Pharmacochem. Libr. 16,213-216. 

The artificial neural network (ANN) is a system imitating the operation of a biological neural network. It is composed of the set of basic elements (artificial neurons) that are mutually connected. In general, to describe the ANN operation at least three basic properties should be known namely a neuron model (transfer function), the network topology and the method of training. 

---