Complex-Valued Neural Networks: A New Learning Strategy Using Particle Swarm Optimization

In this chapter, the authors will try to go through the problem of learning the complex-valued neural networks (CVNNs) using particle swarm optimization (PSO); which is one of the open topics in the machine learning society. Quantitative structure-activity relationship (QSAR) modelling is one of the well developed areas in drug development through computational chemistry. This relationship between molecular structure and change in biological activity is center of focus for QSAR modelling. Machine learning algorithms are important tools for QSAR analysis, as a result, they are integrated into the drug production process. Predicting the real-valued drug activity problem is modelled by the CVNN and is learned by a new strategy based on PSO. The trained CVNNs are tested on two drug sets as a real world bench-mark problem. The results show that the prediction and generalization abilities of CVNNs is superior in comparison to the conventional real-valued neural networks (RVNNs). Moreover, convergence of CVNNs is much faster than that of RVNNs in most of the cases. Complex-Valued Neural Networks: A New Learning Strategy Using Particle Swarm Optimization

[1]  Walter Cedeño,et al.  A comparison of particle swarms techniques for the development of quantitative structure-activity relationship models for drug design , 2005, 2005 IEEE Computational Systems Bioinformatics Conference - Workshops (CSBW'05).

[2]  C. Hansch.  Quantitative approach to biochemical structure-activity relationships , 1969 .

[3]  Tohru Nitta,et al.  An Extension of the Back-Propagation Algorithm to Complex Numbers , 1997, Neural Networks.

[4]  Necati Özdemir,et al.  Complex valued neural network with Möbius activation function , 2011 .

[5]  A. Hirose.  Nature of complex number and complex-valued neural networks , 2011 .

[6]  Claudio Moraga,et al.  Multilayer Feedforward Neural Network Based on Multi-valued Neurons (MLMVN) and a Backpropagation Learning Algorithm , 2006, Soft Comput..

[7]  Kazuyuki Murase,et al.  Single-layered complex-valued neural network for real-valued classification problems , 2009, Neurocomputing.

[8]  J C Gertrudes,et al.  Machine learning techniques and drug design. , 2012, Current medicinal chemistry.

[9]  Włodzisław Duch,et al.  Artificial intelligence approaches for rational drug design and discovery. , 2007, Current pharmaceutical design.

[10]  John Steele,et al.  Drug-like properties: guiding principles for design - or chemical prejudice? , 2004, Drug discovery today. Technologies.

[11]  Akira Hirose,et al.  Generalization Characteristics of Complex-Valued Feedforward Neural Networks in Relation to Signal Coherence , 2012, IEEE Transactions on Neural Networks and Learning Systems.

[12]  C. Lipinski.  Lead- and drug-like compounds: the rule-of-five revolution. , 2004, Drug discovery today. Technologies.

[13]  Eduardo A. Castro,et al.  Computer-Aided Linear Modeling Employing Qsar for Drug Discovery , 2009 .

[14]  D. Agrafiotis,et al.  Feature selection for structure-activity correlation using binary particle swarms. , 2002, Journal of medicinal chemistry.

[15]  J. Dearden,et al.  Linear QSAR regression models for the prediction of bioconcentration factors by physicochemical properties and structural theoretical molecular descriptors. , 2007, Chemosphere.

[16]  Masaki Kobayashi,et al.  Exceptional Reducibility of Complex-Valued Neural Networks , 2010, IEEE Transactions on Neural Networks.

[17]  Catherine Blake,et al.  UCI Repository of machine learning databases , 1998 .

[18]  James Kennedy,et al.  Particle swarm optimization , 2002, Proceedings of ICNN'95 - International Conference on Neural Networks.

[19]  Mohammed E. El-Telbany,et al.  The Predictive Learning Role in Drug Design , 2014 .

[20]  Albert P. Li,et al.  Preclinical in vitro screening assays for drug-like properties. , 2005, Drug discovery today. Technologies.