Learning soft computing control strategies in a modular neural network architecture

Modelling and control of nonlinear dynamical systems is a challenging problem since the dynamics of such systems change over their parameter space. Conventional methodologies for designing nonlinear control laws, such as gain scheduling, are effective because the designer partitions the overall complex control into a number of simpler sub-tasks. This paper describes a new genetic algorithm based method for the design of a modular neural network (MNN) control architecture that learns such partitions of an overall complex control task. Here a chromosome represents both the structure and parameters of an individual neural network in the MNN controller and a hierarchical fuzzy approach is used to select the chromosomes required to accomplish a given control task. This new strategy is applied to the end-point tracking of a single-link flexible manipulator modelled from experimental data. Results show that the MNN controller is simple to design and produces superior performance compared to a single neural network (SNN) controller which is theoretically capable of achieving the desired trajectory. (C) 2003 Elsevier Ltd. All rights reserved.

Improving neural network training solutions using regularisation

This paper describes the application of regularisation to the training of feedforward neural networks, as a means of improving the quality of solutions obtained. The basic principles of regularisation theory are outlined for both linear and nonlinear training and then extended to cover a new hybrid training algorithm for feedforward neural networks recently proposed by the authors. The concept of functional regularisation is also introduced and discussed in relation to MLP and RBF networks. The tendency for the hybrid training algorithm and many linear optimisation strategies to generate large magnitude weight solutions when applied to ill-conditioned neural paradigms is illustrated graphically and reasoned analytically. While such weight solutions do not generally result in poor fits, it is argued that they could be subject to numerical instability and are therefore undesirable. Using an illustrative example it is shown that, as well as being beneficial from a generalisation perspective, regularisation also provides a means for controlling the magnitude of solutions. (C) 2001 Elsevier Science B.V. All rights reserved.

Artificial Intelligence Model for Flowable Concrete Mixtures Used in Underwater Construction and Repair

This study explores using artificial neural networks to predict the rheological and mechanical properties of underwater concrete (UWC) mixtures and to evaluate the sensitivity of such properties to variations in mixture ingredients. Artificial neural networks (ANN) mimic the structure and operation of biological neurons and have the unique ability of self-learning, mapping, and functional approximation. Details of the development of the proposed neural network model, its architecture, training, and validation are presented in this study. A database incorporating 175 UWC mixtures from nine different studies was developed to train and test the ANN model. The data are arranged in a patterned format. Each pattern contains an input vector that includes quantity values of the mixture variables influencing the behavior of UWC mixtures (that is, cement, silica fume, fly ash, slag, water, coarse and fine aggregates, and chemical admixtures) and a corresponding output vector that includes the rheological or mechanical property to be modeled. Results show that the ANN model thus developed is not only capable of accurately predicting the slump, slump-flow, washout resistance, and compressive strength of underwater concrete mixtures used in the training process, but it can also effectively predict the aforementioned properties for new mixtures designed within the practical range of the input parameters used in the training process with an absolute error of 4.6, 10.6, 10.6, and 4.4%, respectively.

Partial agonism, neutral antagonism, and inverse agonism at the human wild-type and constitutively active cholecystokinin-2 receptors

Cholecystokinin receptor-2 (CCK2R) is a G protein receptor that regulates a number of physiological functions. Activation of CCK2R and/or expression of a constitutively active CCK2R variant may contribute to human diseases, including digestive cancers. Search for antagonists of the CCK2R has been an important challenge during the last few years, leading to discovery of a set of chemically distinct compounds. However, several early-discovered antagonists turned out to be partial agonists. In this context, we carried out pharmacological characterization of six CCK2R antagonists using COS-7 cells expressing the human CCK2R or a CCK2R mutant having a robust constitutive activity on inositol phosphates production, and we investigated the molecular mechanisms which, at a CCK2R binding site, account for these features. Results indicated that three compounds, 3R(+)-N-(2,3-dihydro-1-methyl-2-oxo-5-phenyl-1H-1,4-benzodiazepin-3- yl)-N'-(3-methylphenyl)urea (L365,260), 4-{[2-[[3-(lH-indol-3-yl)-2- methyl-1-oxo-2-[[[1.7.7-trimethyl-bicyclo[2.2.1]hept-2-yl)-oxy]carbonyl]amino] propyl]amino]-1-phenylethyl]amino-4-oxo-[lS-la.2[S*(S*)]4a]} -butanoate N-methyl-D-glucamine (PD135, 158), and (R)-1-naphthalenepropanoic acid, b-[2-[[2-(8-azaspiro-[4.5]dec-8-ylcarbonyl)-4,6-dimethylphenyl]amino]-2- oxoethyl] (CR2945), were partial agonists; one molecule, 1-[(R)-2,3-dihydro-1- (2,3-dihydro-1-(2-methylphenacyl)-2-oxo-5-phenyl-1H-1,4-benzodiazepin-3-yl] -3-(3-methylphenyl)urea (YM022), was a neutral antagonist; and two compounds, N-(+)-[1-(adamant-1-ylmethyl)-2,4-dioxo-5-phenyl2,3,4,5-tetrahydro-1H-1, 5-benzodiazepin-3-yl]-N'-phenylurea (GV150,013X) and ([(N-[methoxy-3 phenyl] N-[N-methyl N-phenyl carbamoylmethyl], carbomoylmethyl)-3 ureido]-3-phenyl)2-propionic acid (RPR101,048), were inverse agonists. Furthermore, target- and pharmacophore-based docking of ligands followed by molecular dynamic simulation experiments resulted in consistent motion of aromatic residues belonging to a network presumably important for activation, thus providing the first structural explanations for the different pharmacological profiles of tested compounds. This study confirms that several referenced so-called antagonists are in fact partial agonists, and because of this undesired activity, we suggest that newly generated molecules should be preferred to efficiently block CCK2R-related physiological effects. Furthermore, data on the structural basis for the different pharmacological features of CCK2R ligands will serve to further clarify CCK2R mechanism of activation. Copyright © 2006 The American Society for Pharmacology and Experimental Therapeutics.

A fast multi-output RBF neural network construction method

This paper investigates the center selection of multi-output radial basis function (RBF) networks, and a multi-output fast recursive algorithm (MFRA) is proposed. This method can not only reveal the significance of each candidate center based on the reduction in the trace of the error covariance matrix, but also can estimate the network weights simultaneously using a back substitution approach. The main contribution is that the center selection procedure and the weight estimation are performed within a well-defined regression context, leading to a significantly reduced computational complexity. The efficiency of the algorithm is confirmed by a computational complexity analysis, and simulation results demonstrate its effectiveness. (C) 2010 Elsevier B.V. All rights reserved.

ARE ARTIFICIAL NEURAL NETS AS SMART AS A RAT

The origins of artificial neural networks are related to animal conditioning theory: both are forms of connectionist theory, which in turn derives from the empiricist philosophers' principle of association. The parallel between animal learning and neural nets suggests that interaction between them should benefit both sides.

A multiagent network for peer norm enforcement

In a multiagent system where norms are used to regulate the actions agents ought to execute, some agents may decide not to abide by the norms if this can benefit them. Norm enforcement mechanisms are designed to counteract these benefits and thus the motives for not abiding by the norms. In this work we propose a distributed mechanism through which agents in the multiagent system that do not abide by the norms can be ostracised by their peers. An ostracised agent cannot interact anymore and looses all benefits from future interactions. We describe a model for multiagent systems structured as networks of agents, and a behavioural model for the agents in such systems. Furthermore, we provide analytical results which show that there exists an upper bound to the number of potential norm violations when all the agents exhibit certain behaviours. We also provide experimental results showing that both stricter enforcement behaviours and larger percentage of agents exhibiting these behaviours reduce the number of norm violations, and that the network topology influences the number of norm violations. These experiments have been executed under varying scenarios with different values for the number of agents, percentage of enforcers, percentage of violators, network topology, and agent behaviours. Finally, we give examples of applications where the enforcement techniques we provide could be used.

Loading Artificial Magnetic Conductor and Artificial Magnetic Conductor Absorber with Negative Impedance Convertor Elements

The general properties of a frequency selective surface loaded with negative impedance converter (NIC)-based active loads are discussed from a theoretical perspective.The stability problem associated with NIC circuits embedded in artificial magnetic conductor (AMC) and AMC absorber applications is studied using pole-zero analysis. The requirements and constraints for achieving stable operation with enhanced bandwidth using negative capacitance as realized by a floating NIC network are derived. Furthermore, it is shown that it is nearly impossible to simultaneously implement a negative capacitor and a negative inductor to such structures. © 2012 Wiley Periodicals, Inc. Microwave Opt Technol Lett 54:2111–2114, 2012; View this article online at wileyonlinelibrary.com. DOI 10.1002/mop.27019

METAMORPHOSE VI – the Virtual Institute for artificial electromagnetic materials and metamaterials: Origin, mission, and activities

In this paper, we outline the background, mission, and activities of the Virtual Institute for Artificial Electromagnetic Materials and Metamaterials (METAMORPHOSE VI). This international association, founded in the framework of the FP-6 Network of Excellence METAMORPHOSE, aims at promoting and developing research, training, and dissemination activities in the emerging and highly dynamic field of advanced electromagnetic materials and metamaterials at both European and International levels. More than 300 researchers are currently associated with the METAMORPHOSE VI which networks them together in a learnt society. After a brief description of the association and its mission, we present an overview of the activities developed by the METAMORPHOSE VI, with a particular emphasis on the coordination of the European Doctoral Program on Metamaterials (EUPROMETA) and the organization of the International Congress on Advanced Electromagnetic Materials in Microwaves and Optics – metamaterials congress.

Modeling of titanium alloys by using artificial neural networks

Titanium alloy exhibits an excellent combination of bio-compatibility, corrosion resistance, strength and toughness. The microstructure of an alloy influences the properties. The microstructures depend mainly on alloying elements, method of production, mechanical, and thermal treatments. The relationships between these variables and final properties of the alloy are complex, non-linear in nature, which is the biggest hurdle in developing proper correlations between them by conventional methods. So, we developed artificial neural networks (ANN) models for solving these complex phenomena in titanium alloys.

In the present work, ANN models were used for the analysis and prediction of the correlation between the process parameters, the alloying elements, microstructural features, beta transus temperature and mechanical properties in titanium alloys. Sensitivity analysis of trained neural network models were studied which resulted a better understanding of relationships between inputs and outputs. The model predictions and the analysis are well in agreement with the experimental results. The simulation results show that the average output-prediction error by models are less than 5% of the prediction range in more than 95% of the cases, which is quite acceptable for all metallurgical purposes.

Two-Stage Orthogonal Least Squares Methods for Neural Network Construction

A number of neural networks can be formulated as the linear-in-the-parameters models. Training such networks can be transformed to a model selection problem where a compact model is selected from all the candidates using subset selection algorithms. Forward selection methods are popular fast subset selection approaches. However, they may only produce suboptimal models and can be trapped into a local minimum. More recently, a two-stage fast recursive algorithm (TSFRA) combining forward selection and backward model refinement has been proposed to improve the compactness and generalization performance of the model. This paper proposes unified two-stage orthogonal least squares methods instead of the fast recursive-based methods. In contrast to the TSFRA, this paper derives a new simplified relationship between the forward and the backward stages to avoid repetitive computations using the inherent orthogonal properties of the least squares methods. Furthermore, a new term exchanging scheme for backward model refinement is introduced to reduce computational demand. Finally, given the error reduction ratio criterion, effective and efficient forward and backward subset selection procedures are proposed. Extensive examples are presented to demonstrate the improved model compactness constructed by the proposed technique in comparison with some popular methods.

Approximating Credal Network Inferences by Linear Programming

An algorithm for approximate credal network updating is presented. The problem in its general formulation is a multilinear optimization task, which can be linearized by an appropriate rule for fixing all the local models apart from those of a single variable. This simple idea can be iterated and quickly leads to very accurate inferences. The approach can also be specialized to classification with credal networks based on the maximality criterion. A complexity analysis for both the problem and the algorithm is reported together with numerical experiments, which confirm the good performance of the method. While the inner approximation produced by the algorithm gives rise to a classifier which might return a subset of the optimal class set, preliminary empirical results suggest that the accuracy of the optimal class set is seldom affected by the approximate probabilities