Neurocomputing Research Papers - Academia.edu (original) (raw)

The adequacy of radial basis function neural networks to model the inside air temperature of a hydroponic greenhouse as a function of the outside air temperature and solar radiation, and the inside relative humidity, is addressed. As the model is intended to be incorporated in an environmental control strategy both o -line and on-line methods could be of use to accomplish this task. In this paper known hybrid o -line training methods and on-line learning algorithms are analyzed. An o -line method and its application to on-line learning is proposed. It exploits the linear-non-linear structure found in radial basis function neural networks.

- by
- •
- Engineering, Neural Networks, Environmental Control, Radial Basis Functions

An evolutive algorithm for the optimal design of wind farms is presented. The algorithm objective is to optimize the profits given an investment on a wind farm. Net present value will be used as a figure of the revenue. To work out this figure, several economic factors such as the initial capital investment, the discount rate, the value of the generated energy and the number of the years spanned by the investment are considered. All this factors depends on the preliminary design of the wind park (number, type, tower height and layout situation of wind generators), which are the variables to set. r

- by Castro Mora
- •
- Engineering, Genetic Algorithm, Optimization, Capital Investment

Current models of cortical computation are based on analog quantities instead of single spikes. This paper extends the predictive coding model (Nature Neurosci. 2(1) (1999) 79) to the level of neural signaling. Neurons in our model use a mixed strategy to transmit information. Spikes are not only messages of computation, but also carriers of information with analog quantities encoded in their phases. Computation is shared among cells both in time and in space, such that information is signaled probabilistically in a distributed synchronous fashion. Contrary to "noise other than signal" interpretation of irregularity of neural signaling, our model proposes a computational role of such variability.

- by zuohua zhang
- •
- Engineering, Predictive coding, Minimum description length, Neurocomputing

In this paper, a novel quantum swarm evolutionary algorithm (QSE) is presented based on the quantum-inspired evolutionary algorithm (QEA). A new definition of Q-bit expression called quantum angle is proposed, and an improved particle swarm optimization (PSO) is employed to update the quantum angles automatically. The simulated results in solving 0-1 knapsack problem show that QSE is superior to traditional QEA. In addition, the comparison experiments show that QSE is better than many traditional heuristic algorithms, such as climb hill algorithm, simulation anneal algorithm and taboo search algorithm. Meanwhile, the experimental results of 14 cities traveling salesman problem (TSP) show that it is feasible and effective for small-scale TSPs, which indicates a promising novel approach for solving TSPs. r

- by Yan Liang
- •
- Engineering, Particle Swarm Optimization, Search Algorithm, Traveling Salesman Problem

An application of the self-organizing map (SOM) to the Traveling Salesman Problem (TSP) has been reported by many researchers, however these approaches are mainly focused on the Euclidean TSP variant. We consider the TSP as a problem formulation for the multi-goal path planning problem in which paths among obstacles have to be found. We apply a simple approximation of the shortest path that seems to be suitable for the SOM adaptation procedure. The approximation is based on a geometrical interpretation of SOM, where weights of neurons represent nodes that are placed in the polygonal domain. The approximation is verified in a set of real problems and experimental results show feasibility of the proposed approach for the SOM based solution of the non-Euclidean TSP.

- by Jan Faigl
- •
- Engineering, Neural Network, Path planning, Traveling Salesman Problem

This paper introduces a novel recognition framework for human actions using hybrid features. The hybrid features consist of spatio-temporal and local static features extracted using motion-selectivity attribute of 3D dual-tree complex wavelet transform (3D DT-CWT) and affine SIFT local image detector, respectively. The proposed model offers two core advantages: (1) the framework is significantly faster than traditional approaches due to volumetric processing of images as a '3D box of data' instead of a frame by frame analysis, (2) rich representation of human actions in terms of reduction in artifacts in view of the promising properties of our recently designed full symmetry complex filter banks with better directionality and shift-invariance properties. No assumptions about scene background, location, objects of interest, or point of view information are made whereas bidirectional two-dimensional PCA (2D-PCA) is employed for dimensionality reduction which offers enhanced capabilities to preserve structure and correlation amongst neighborhood pixels of a video frame.

- by Rashid Minhas
- •
- Engineering, Activity Recognition, Action Recognition, Complex wavelet transform

Several approaches appear in literature in order to develop Computed-Aided-Diagnosis (CAD) systems for Alzheimer's disease (AD) detection. Although univariate models became very popular and nowadays they are widely used, recent investigations are focused on multivariate models which deal with a whole image as an observation. In this work, we compare two multivariate approaches that use different methodologies to relieve the small sample size problem. One of them is based on Gaussian Mixture Model (GMM) and models the Regions of Interests (ROIs) defined as differences between controls and AD subject. After GMM estimation using the EM algorithm, feature vectors are extracted for each image depending on the positions of the resulting Gaussians. The other method under study computes score vectors through a Partial Least Squares (PLS) algorithm based estimation and those vectors are used as features. Before extracting the score vectors, a binary mask based dimensional reduction of the input space is performed in order to remove low-intensity voxels. The validity of both methods is tested on the ADNI database by implementing several CAD systems with linear and nonlinear classifiers and comparing them with previous approaches such as VAF and PCA. (J.M. Gó rriz). 1 Data used in preparation of this article were obtained from the Alzheimer's disease Neuroimaging Initiative (ADNI) database (www.loni.ucla.edu/ADNI). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. ADNI investigators include (complete listing available at

- by Oscar Castillo
- •
- Engineering, Neurocomputing

The ability to forecast the future based on past data is a key tool to support individual and organizational decision making. In particular, the goal of Time Series Forecasting (TSF) is to predict the behavior of complex systems by looking only at past patterns of the same phenomenon. In recent years, several works in the literature have adopted Evolutionary Artificial Neural Networks (EANN) for TSF. In this work, we propose a novel EANN approach, where a weighted n-fold validation fitness scheme is used to build an ensemble of neural networks, under four different combination methods: mean, median, softmax and rank-based. Several experiments were held, using six real-world time series with different characteristics and from distinct domains. Overall, the proposed approach achieved competitive results when compared with a non-weighted n-fold EANN ensemble, the simpler 0-fold EANN and also the popular Holt-Winters statistical method.

- by Araceli Sanchis de Miguel
- •
- Engineering, Mathematics, Computer Science, Evolutionary Computation

In order to facilitate complexity optimization in feedforward networks, several algorithms are developed that combine growing and pruning. First, a growing scheme is presented which iteratively adds new hidden units to full-trained networks. Then, a non-heuristic onepass pruning technique is presented, which utilizes orthogonal least squares. Based upon pruning, a one-pass approach is developed for generating the validation error versus network size curve. A combined approach is described in which networks are continually pruned during the growing process. As a result, the hidden units are ordered according to their usefulness, and the least useful units are eliminated. Examples show that networks designed using the combined method have less training and validation error than growing or pruning alone. The combined method exhibits reduced sensitivity to the initial weights and generates an almost monotonic error versus network size curve. It is shown to perform better than two well-known growing methods-constructive backpropagation and cascade correlation. r

- by Michael Manry
- •
- Engineering, Back Propagation, Network Design, Neurocomputing

A consortium of nine European industrial enterprises and research institutions from four states is attempting to discover the industrial potential for application of neural networks. This technology is being exploited by over 100 companies in the USA. The ANNIE project aims to help European companies to exploit neural network principles profitably. It is intended to find the types of applications in which neural networks offer improved technical solutions in comparison to or in combination with conventional approaches. A handbook will be prepared which shows the features of problems where neural networks have something to offer. ANNIE will increasingly become concerned with demonstrating prototype applications and with the need for transparent interfaces for the end-user.

- by Andrew Chadwick
- •
- Engineering, Neural Network, Neurocomputing

A fast learning neuroevolutionary algorithm for both feedforward and recurrent networks is proposed. The method is inspired by the well known and highly effective Cartesian genetic programming (CGP) technique. The proposed method is called the CGP-based Artificial Neural Network (CGPANN). The basic idea is to replace each computational node in CGP with an artificial neuron, thus producing an artificial neural network. The capabilities of CGPANN are tested in two diverse problem domains. Firstly, it has been tested on a standard benchmark control problem: single and double pole for both Markovian and non-Markovian cases. Results demonstrate that the method can generate effective neural architectures in substantially fewer evaluations in comparison to previously published neuroevolutionary techniques. In addition, the evolved networks show improved generalization and robustness in comparison with other techniques. Secondly, we have explored the capabilities of CGPANNs for the diagnosis of Breast Cancer from the FNA (Finite Needle Aspiration) data samples. The results demonstrate that the proposed algorithm gives 99.5% accurate results, thus making it an excellent choice for pattern recognitions in medical diagnosis, owing to its properties of fast learning and accuracy.

- by Maryam Khan
- •
- Engineering, Neuroevolution, Neurocomputing, Artificial Neural Network

Blind source separation problems have recently drawn a lot of attention in unsupervised neural learning. In the current approaches, the number of sources is typically assumed to be known in advance, but this does not usually hold in practical applications. In this paper, various neural network architectures and associated adaptive learning algorithms are discussed for handling the cases where the number of sources is unknown. These techniques include estimation of the number of sources, redundancy removal among the outputs of the networks, and extraction of the sources one at a time. Validity and performance of the described approaches are demonstrated by extensive computer simulations for natural image and magnetoencephalographic (MEG) data. 0925-2312/99/$ -see front matter 1999 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 5 -2 3 1 2 ( 9 8 ) 0 0 0 9 1 -5

- by Juha Karhunen
- •
- Engineering, Image Processing, Blind Source Separation, Neural Network

In adversarial classification tasks like spam filtering and intrusion detection, malicious adversaries may manipulate data to thwart the outcome of an automatic analysis. Thus, besides achieving good classification performances, machine learning algorithms have to be robust against adversarial data manipulation to successfully operate in these tasks. While support vector machines (SVMs) have shown to be a very successful approach in classification problems, their effectiveness in adversarial classification tasks has not been extensively investigated yet. In this paper we present a preliminary investigation of the robustness of SVMs against adversarial data manipulation. In particular, we assume that the adversary has control over some training data, and aims to subvert the SVM learning process. Within this assumption, we show that this is indeed possible, and propose a strategy to improve the robustness of SVMs to training data manipulation based on a simple kernel matrix correction.

- by huang xiao and +2
- •
- Engineering, Support Vector Machines, Neurocomputing

Although nonnegative matrix factorization (NMF) favors a sparse and part-based representation of nonnegative data, there is no guarantee for this behavior. Several authors proposed NMF methods which enforce sparseness by constraining or penalizing the ' 1 -norm of the factor matrices. On the other hand, little work has been done using a more natural sparseness measure, the ' 0 -pseudo-norm. In this paper, we propose a framework for approximate NMF which constrains the ' 0 -norm of the basis matrix, or the coefficient matrix, respectively. For this purpose, techniques for unconstrained NMF can be easily incorporated, such as multiplicative update rules, or the alternating nonnegative least-squares scheme. In experiments we demonstrate the benefits of our methods, which compare to, or outperform existing approaches.

- by Petia Georgieva
- •
- Engineering, Gene expression, Cluster Analysis, Feature Extraction
- by Deepika saxena
- •
- Engineering, Neurocomputing, Psychology and Cognitive Sciences

In this paper we present denoising algorithms for enhancing noisy signals based on Local ICA (LICA), Delayed AMUSE (dAMUSE) and Kernel PCA (KPCA). The algorithm LICA relies on applying ICA locally to clusters of signals embedded in a high dimensional feature space of delayed coordinates. The components resembling the signals can be detected by various criteria like estimators of kurtosis or the variance of autocorrelations depending on the statistical nature of the signal. The algorithm proposed can be applied favorably to the problem of denoising multidimensional data. Another projective subspace denoising method using delayed coordinates has been proposed recently with the algorithm dAMUSE. It combines the solution of blind source separation problems with denoising efforts in an elegant way and proofs to be very efficient and fast. Finally, KPCA represents a non-linear projective subspace method that is well suited for denoising also. Besides illustrative applications to toy examples and images, we provide an application of all algorithms considered to the analysis of protein NMR spectra.

- by Ana Teixeira
- •
- Engineering, Blind Source Separation, Neurocomputing, High Dimensionality

Many credit data classification problems require label predictions only for a given unlabeled test set. Since the number of an available unlabeled test data set is much larger than a labeled data set, it is desirable to build a predictive model in a transductive setting that takes advantage of the unlabeled data as well as labeled data. This paper proposes a localized transduction based multi-layer perceptron (MLP) methodology to build a better classifier. We provide a practical framework for our methodology. Simulations on real credit delinquents detection problems are conducted to test the proposed method with a promising result.

- by HYEJIN PARK
- •
- Engineering, Multi-layer Perceptron, Transduction, Prediction Model

Using class label fuzzification, this study develops the idea of refreshing the attitude of the difficult training examples and gaining a more robust classifier for large-margin support vector machines (SVMs). Fuzzification relaxes the specific hard-limited Lagrangian constraints of the difficult examples, extends the infeasible space of the canonical constraints for optimization, and reconfigures the consequent decision function with a wider margin. With the margin, a classifier capable of achieving a high generalization performance can be more robust. This paper traces the rationale for such a robust performance back to the changes of governing loss function. From the aspect of loss function, the reasons are causally explained. In the study, we also demonstrate a two-stage system for experiments to show the changes corresponding to the label fuzzification. The system first captures the difficult examples in the first-stage preprocessor, and assigns them various fuzzified class labels. Three types of membership functions, including a constant, a linear, and a sigmoidal membership function, are designated in the preprocessor to manipulate the within-class correlations of the difficult examples for reference of the fuzzification. The consequent performance benchmarks confirm the robust and generalized ability due to the label fuzzification. Since the change of y 0 i is fundamental, the idea may be transplanted to different prototypes of SVM.

- by Jui Jen Chou
- •
- Engineering, Mathematics, Computer Science, Machine Learning
- by Benjamin Schrauwen
- •
- Engineering, Excitable Media, Recurrent Neural Network, Reservoir Computing

In this paper an effective dynamic video summarisation algorithm is presented using audio-visual features extracted from videos. Audio, colour and motion features are dynamically fused using an adaptively weighting mechanism. Dissimilarities of temporal video segments are formulated using the extracted features before these segments are clustered using a fuzzy c-means algorithm with an optimally determined cluster number. The experimental results demonstrate the ability of the proposed algorithm to automatically summarise the videos with good performance.

- by Jawid Azizi
- •
- Engineering, Video segmentation, Audio Visual, Feature Extraction

Currently, cancer diagnosis at a molecular level has been made possible through the analysis of gene expression data. More specifically, one usually uses machine learning (ML) techniques to build, from cancer gene expression data, automatic diagnosis models (classifiers). Cancer gene expression data often present some characteristics that can have a negative impact in the generalization ability of the classifiers generated. Some of these properties are data sparsity and an unbalanced class distribution. We investigate the results of a set of indices able to extract the intrinsic complexity information from the data. Such measures can be used to analyze, among other things, which particular characteristics of cancer gene expression data mostly impact the prediction ability of support vector machine classifiers. In this context, we also show that, by applying a proper feature selection procedure to the data, one can reduce the influence of those characteristics in the error rates of the classifiers induced.

- by getahun kassie
- •
- Information Systems, Engineering, Mechanical Engineering, Applied Mathematics

This work presents a new prediction-based portfolio optimization model that can capture short-term investment opportunities. We used neural network predictors to predict stocks' returns and derived a risk measure, based on the prediction errors, that have the same statistical foundation of the mean-variance model. The efficient diversification effects holds thanks to the selection of predictors with low and complementary pairwise error profiles.

- by Alberto Ferreira De Souza
- •
- Engineering, Time Series, Neural Networks, Neural Network

The input-output properties of motoneurons (MNs) and motor units (MUs) may be modulated by different physiological variables, including neuromodulators released by presynaptic neurons from the brainstem. Monoamines, such as serotonin and norepinephrine, act on MNs mainly by activating dendritic L-type Ca þ þ channels, generating a persistent inward current (PIC), which may change the input-output properties of the MU. If the firing properties of individual MNs and also the features of the force generated by the MUs can be changed, it follows that the descending monoaminergic systems may modulate the overall dynamics of motor control. The purpose of the present study is to investigate the effects of neuromodulation on the input-output properties of mathematically modeled typespecified MUs. Computationally efficient models are presented for S-and F-type mammalian MNs as well as a MN pool commanding muscle units of the Soleus muscle. The single models have a dendritic L-type Ca þ þ channel, generating a PIC, along with somatic currents that are responsible for spikegenerating mechanisms and the afterhyperpolarization. The S-type active-dendrite (AD) MN model resulted highly excitable and discharged in a self-sustained manner after an excitatory input (bistability). An inhibitory activity turned off this self-sustained discharge. In addition, this lowthreshold MN model showed a significant reduction in interspike interval (ISI) variability in comparison with an equivalent passive-dendrite (PD) MN model discharging at a similar mean rate. The frequency response gain from presynaptic spike train rate modulation to output spike train modulation had a clear valley from about 1-10 Hz for the S-type AD MN model, whereas the PD model showed a gain increase instead. On the other hand, the frequency responses of PD and AD F-type models were similarly shaped, with the AD model having a higher gain at high frequencies. These results suggest that in motor behaviors where steady or low frequency activity is required, such as posture, PICs would aid low-threshold MNs to respond with regular spikes, reducing output variability, and would attenuate the effects of high-frequency input disturbances, helping maintain system steadiness. At the same time, high-threshold MNs being more responsive to high-frequency disturbances would contribute with the necessary activity to correct for postural deviations from a desired position. Simulation results from the neuromuscular model have shown that the PIC activation profoundly affects muscle force generation, with the neuromodulatory activity acting in the adjustment of the motor output. Both individual and collective results presented here expand the understanding of the versatility of monoaminergic neuromodulation in adjusting both the MN input-output coding and the control of force generation.

- by André Kohn and +1
- •
- Engineering, Motor Control, ISI, Neuronal Network

Hardware FPGA MACs a b s t r a c t Artificial neural networks (ANNs) is a well known bio-inspired model that simulates human brain capabilities such as learning and generalization. ANNs consist of a number of interconnected processing units, wherein each unit performs a weighted sum followed by the evaluation of a given activation function. The involved computation has a tremendous impact on the implementation efficiency. Existing hardware implementations of ANNs attempt to speed up the computational process. However these implementations require a huge silicon area that makes it almost impossible to fit within the resources available on a state-of-the-art FPGAs. In this paper, we devise a hardware architecture for ANNs that takes advantage of the dedicated adder blocks, commonly called MACs to compute both the weighted sum and the activation function. The proposed architecture requires a reduced silicon area considering the fact that the MACs come for free as these are FPGA's built-in cores. The hardware is as fast as existing ones as it is massively parallel. Besides, the proposed hardware can adjust itself on-thefly to the user-defined topology of the neural network, with no extra configuration, which is a very nice characteristic in robot-like systems considering the possibility of the same hardware may be exploited in different tasks.

- by Nadia Nedjah
- •
- Engineering, Neural Network, Human Brain, Neurocomputing

Bankruptcy prediction has been widely studied as a binary classification problem using financial ratios methodologies. In this paper, Leave-One-Out-Incremental Extreme Learning Machine (LOO-IELM) is explored for this task. LOO-IELM operates in an incremental way to avoid inefficient and unnecessary calculations and stops automatically with the neurons of which the number is unknown. Moreover, Combo method and further Ensemble model are investigated based on different LOO-IELM models and the specific financial indicators. These indicators are chosen using different strategies according to the financial expertise. The entire process has shown its good performance with a very fast speed, and also helps to interpret the model and the special ratios.

- by Yoan Miche
- •
- Engineering, Neurocomputing

A recently-developed mathematical semantic theory explains the relationship between knowledge and its representation in connectionist systems. The semantic theory is based upon category theory, the mathematical theory of structure. A product of its explanatory capability is a set of principles to guide the design of future neural architectures and enhancements to existing designs. We claim that this mathematical semantic approach to network design is an effective basis for advancing the state of the art. We offer two experiments to support this claim. One of these involves multispectral imaging using data from a satellite camera. 2 weights? How do we understand connectionist learning, generalization, and specialization in terms of data and prior knowledge? A systematic means of addressing these questions can serve as a fundamental and comprehensive base for analysis and design provided that the understanding is accompanied by mathematical rigor. But this begs the question of whether there is a form of mathematics that can serve as a vehicle for addressing questions about neural network semantics. We introduce a form of mathematics that serves as such a vehicle along with an experiment testing this claim.

- by Kurt Larson and +1
- •
- Engineering, Multispectral Imaging, Category Theory, Network Design

Perspiration phenomenon is very significant to detect liveness of a finger. However, it requires two consecutive fingerprints to notice perspiration, and therefore it may not be suitable for real-time authentications. Some other methods in the literature need extra hardware to detect liveness. To alleviate these problems, we propose a new ridgelet transform-based method which needs only one fingerprint to detect liveness. Wavelets are very effective in representing objects with isolated point singularities, but failed to represent line singularities. Ridgelet transform allows representing singularities along lines in a more efficient way than the wavelets. Fingerprint is an oriented texture pattern of ridge lines; hence naturally ridgelets are more suitable for fingerprint processing than the wavelets. We use ridgelet energy and co-occurrence signatures to characterize fingerprint texture using our databases consisting of real and spoof fingerprints. Dimensionalities of feature sets are reduced by running principal component analysis (PCA) algorithm. Ridgelet energy and co-occurrence signatures are independently tested on various classifiers such as: neural network, support vector machine and K-nearest neighbor. Finally, we fuse all the classifiers using the ''mean rule'' to build an ensemble classifier. Fingerprint databases consisting of 185 real, 90 fun-doh and 150 gummy fingerprints are created. Multiple combinations of materials are used to create casts and moulds of spoof fingerprints. Experimental results indicate that, the performance of a new liveness detection approach is very promising, as it needs only one fingerprint and no extra hardware to detect vitality.

- by Suneeta Agarwal
- •
- Engineering, Principal Component Analysis, Neural Network, Support vector machine

The proliferation of specialised workshops associated with the Advances in Neural Information Processing series of conferences has been of almost unqualiÿed beneÿt to the scientiÿc endeavour. In keeping with the plurality which has characterised the NIPS community throughout its history, well-chosen workshop themes have nurtured extended interdisciplinary discussion of issues in neural computation, providing just the sort of intellectual sca olding necessary if the interchange of ideas is to deliver upon its considerable promise.

- by David Saad
- •
- Engineering, Theory and Practice, Neurocomputing

Debutanizer column is an important unit operation in petroleum refining industries. The design of online composition prediction by using neural network will help improve product quality monitoring in an oil refinery industry by predicting the top and bottom composition of n-butane simultaneously and accurately for the column. The single dynamic neural network model can be used and designed to overcome the delay introduced by lab sampling and can be also suitable for monitoring purposes. The objective of this work is to investigate and implement an artificial neural network (ANN) for composition prediction of the top and bottom product of a distillation column simultaneously. The major contribution of the current work is to develop these composition predictions of n-butane by using equation based neural network (NN) models. The composition predictions using this method is compared with partial least square (PLS) and regression analysis (RA) methods to show its superiority over these other conventional methods. Based on statistical analysis, the results indicate that neural network equation, which is more robust in nature, predicts better than the PLS equation and RA equation based methods.

- by Bawadi Abdullah
- •
- Engineering, Neurocomputing

There exists a wide range of paradigms, and a high number of different methodologies that are applied to the problem of time series prediction. Most of them are presented as a modified function approximation problem using input/output data, in which the input data are expanded using values of the series at previous steps. Thus, the model obtained normally predicts the value of the series at a time ðt þ hÞ using previous time steps ðt À t 1 Þ; ðt À t 2 Þ; . . . ; ðt À t n Þ. Nevertheless, learning a model for long term time series prediction might be seen as a more complicated task, since it might use its own outputs as inputs for long term prediction (recursive prediction). This paper presents the utility of two different methodologies, the TaSe fuzzy TSK model and the least-squares SVMs, to solve the problem of long term time series prediction using recursive prediction. This work also introduces some techniques that upgrade the performance of those advanced one-step-ahead models (and in general of any one-step-ahead model), where they are used recursively for long term time series prediction. r

- by Alberto Valenzuela
- •
- Engineering, Time Series Prediction, Neurocomputing, Fuzzy System

Knee contact pressure is a crucial factor in the knee rehabilitation programs. Although contact pressure can be estimated using finite element analysis, this approach is generally time-consuming and does not satisfy the real-time requirements of a clinical set-up. Therefore, a real-time surrogate method to estimate the contact pressure would be advantageous.

- by Mehran Moazen and +1
- •
- Engineering, Neurocomputing
- by Ram Shanmugam
- •
- Engineering, Mathematics, Computer Science, Economics

Data-centric methods like soft computing and machine learning have gained greater interest and acceptance in the oil and gas industry in recent years. We give an overview of the opportunities and challenges facing applied time series prediction in this domain, with a focus on fault prediction. In particular, we argue that the physical processes and hierarchies of information flow in the industry strongly determine the choice of soft computing or machine learning methods.

- by Roar Nybø
- •
- Engineering, Machine Learning, Time Series, Soft Computing

The last 50 years has witnessed considerable research in the area of neural networks resulting in a range of architectures, learning algorithms and demonstrative applications. A more recent research trend has focused on the biological plausibility of such networks as a closer abstraction to real neurons may offer improved performance in an adaptable, real-time environment. This poses considerable challenges for engineers particularly in terms of the requirement to realise a low-cost embedded solution. Programmable hardware has been widely recognised as an ideal platform for the adaptable requirements of neural networks and there has been considerable research reported in the literature. This paper aims to review this body of research to identify the key lessons learned and, in particular, to identify the remaining challenges for large-scale implementations of spiking neural networks on FPGAs. r

- by Jim Harkin
- •
- Engineering, Neural Network, Field Programmable Gate Array, Real Time

This paper presents the use of an intelligent hybrid stock trading system that integrates neural networks, fuzzy logic, and genetic algorithms techniques to increase the efficiency of stock trading when using a volume adjusted moving average (VAMA), a technical indicator developed from equivolume charting. For this research, a neuro-fuzzy-based genetic algorithm (NF-GA) system utilizing a VAMA membership function is introduced. The results show that the intelligent hybrid system takes advantage of the synergy among these different techniques to intelligently generate more optimal trading decisions for the VAMA, allowing investors to make better stock trading decisions.

- by David Enke
- •
- Engineering, Fuzzy Logic, Genetic Algorithms, Neural Networks

Photovoltaic (PV) inverters convert DC voltage and current to AC quantities whose magnitude and frequency are controlled to obtain the desired output. While there are plenty of controllers, it is the fuzzy logic controller (FLC) that receives increasing attention. In this study, a novel metaheuristic optimization algorithm known as lightning search algorithm (LSA) is presented for solving the problem of trial and error procedure in obtaining membership functions (MFs) used in the conventional FLCs. The LSA mimics the natural phenomenon of lightning. It is generalized from the mechanism of step leader propagation. The proposed optimization algorithm considers the concept of fast particles known as projectiles. The probabilistic nature and tortuous characteristics of lightning discharges, which depend on the type of projectile, are modeled using various random distribution functions. To evaluate the reliability and efficiency of the proposed algorithm, the LSA is first tested using 10 benchmark functions with various characteristics necessary to evaluate a new algorithm. Then it is used in designing optimum FLC for standalone PV inverter. The result demonstrates that the LSA generally provides better outcome compared with the other tested methods with a high convergence rate.

- by Mohamed Azah
- •
- Engineering, Neurocomputing

In many applications, data objects are described by both numeric and categorical features. The k-prototype algorithm is one of the most important algorithms for clustering this type of data. However, this method performs hard partition, which may lead to misclassification for the data objects in the boundaries of regions, and the dissimilarity measure only uses the user-given parameter for adjusting the significance of attribute. In this paper, first, we combine mean and fuzzy centroid to represent the prototype of a cluster, and employ a new measure based on co-occurrence of values to evaluate the dissimilarity between data objects and prototypes of clusters. This measure also takes into account the significance of different attributes towards the clustering process. Then we present our algorithm for clustering mixed data. Finally, the performance of the proposed method is demonstrated by a series of experiments on four real world datasets in comparison with that of traditional clustering algorithms.

- by Wei Pang
- •
- Engineering, Data Mining, Fuzzy Clustering, Knowledge Based Systems

We present a novel approach to learn binary classifiers when only positive and unlabeled instances are available (PU learning). This problem is routinely cast as a supervised task with label noise in the negative set. We use an ensemble of SVM models trained on bootstrap subsamples of the training data for increased robustness against label noise. The approach can be considered in a bagging framework which provides an intuitive explanation for its mechanics in a semi-supervised setting. We compared our method to stateof-the-art approaches in a simulation based on the MNIST digit recognition data set. The included benchmark comprises three settings with increasing label noise: (i) fully supervised, (ii) PU learning and (iii) PU learning with false positives. Our approach shows a marginal improvement over existing methods in the second setting and a significant improvement in the third.

- by Johan Suykens
- •
- Engineering, Neurocomputing

how the behaviour of linear control systems with large changes of their parameters might be investigated, • how symmetry might be used to optimize linear control systems, and • how chaotic behaviour might be used in the synthesis of... more

- by Dan Simon
- •
- Engineering, Mechanical Engineering, Applied Mathematics, Fuzzy Logic
- by Navneet Kaur
- •
- Engineering, Cognitive Science, Computer Science, Programming Languages
- by Ian Parberry
- •
- Engineering, Neural Network, Hamiltonian Cycle Problem, Neurocomputing
- by j M
- •
- Engineering, Cognitive Science, Computer Science, Programming Languages

Several novel structure from motion algorithms are presented that are designed to more effectively manage the problem of noise. In many practical applications, structure from motion algorithms fail to work properly because of the noise in the optical flow values. Most structure from motion algorithms implicitly assume that the noise is identically distributed and that the noise is white. Both assumptions are false. Some points can be track more easily than others and some points can be tracked more easily in a particular direction. The accuracy of each optical flow value can be quantified using an optical flow probability distribution. By using optical flow probability distributions in place of optical flow estimates in a structure from motion algorithm, a better understanding of the noise is developed and a more accurate solution is obtained. ix List of Figures 1. Diagram of the Projection onto a unit sphere …………………… 2. Two-Frame Flowchart …………………………………………... 22 3. Multiple-Frame Flowchart ………………………………………. 4. Two-Frame Simulation-Translation …………………………… 5. Two-Frame Simulation-Rotation ……………………………… 36 6. Translation errors for different method that do and do not use probability distributions ...………………………………………. 7. Rotation errors for different method that do and do not use probability distributions ………………………………………… 8. Depth errors for different method that do and do not use probability distributions ……………………………………….. 9. Translation error comparison with Zucchelli's Method ……….. 40 10. Rotation error comparison with Zucchelli's Method …………. 40 11. One frame from a computer-generated video …………………. 12. True inverse depth …………………………………………….. 42 13. Recovered inverse depth using the multiple frames with distributions method …………………………………………… 43 14. Recovered inverse depth using the multiple frames without distributions method …………………………………………... 43 x 15. Recovered inverse depth with the camera moving forward ….. 16. One frame from a video from a camera approaching a tree ….. 43 17. Recovered inverse depth ……………………………………… 43

- by Zhaoyi Wei
- •
- Engineering, Mathematics, Computer Science, Structure from Motion

Connecting the dots between discoveries in neuroscience (neuroplasticity), psychoneuroimmunology(the brain-immune loop) and user experience (gadget rub-off) indicate the nature of our time spent with gadgets is a vector in human health - mentally, socially and physically. The positive design of our interactions with devices therefore can have a positive impact on economy, civilization and society. Likewise, the absence of design that encourages positive interaction may encourage undesirable behaviors. Much like the architecture of physical spaces and buildings, the consequences of the architecture of the 21stcentury conversation between man and machine may last generations. AI and the Internet of Things are primary vectors for positive and negative impacts of technology. We describe a growing body of co-discoveries occurring across a variety of disciplines that support the argument for human sciences in technology design. The paper brings the rigors of science to bear on the humanities and makes the case for its relevance at this juncture in time to the design of technology involved in conversation and behaviors between humans and machines .

- by cindy mason
- •
- Psychology, Cognitive Psychology, Cognitive Science, Social Psychology

The challenge of predicting future values of a time series covers a variety of disciplines. The fundamental problem of selecting the order and identifying the time varying parameters of an autoregressive moving average model (ARMA) concerns many important fields of interest such as linear prediction, system identification and spectral analysis. Recent research activities in forecasting with artificial neural networks (ANNs) suggest that ANNs can be a promising alternative to the traditional ARMA structure. These linear models and ANNs are often compared with mixed conclusions in terms of the superiority in forecasting performance. This study was designed: (a) to investigate a hybrid methodology that combines ANN and ARMA models; (b) to resolve one of the most important problems in time series using ARMA structure and Box-Jenkins methodology: the identification of the model. In this paper, we present a new procedure to predict time series using paradigms such as: fuzzy systems, neural networks and evolutionary algorithms. Our goal is to obtain an expert system based on paradigms of artificial intelligence, so that the linear model can be identified automatically, without the need of human expert participation. The obtained linear model will be combined with ANN, making up an hybrid system that could outperform the forecasting result. r

- by Fernando Rojas
- •
- Engineering, System Identification, Time Series, Soft Computing

This paper addresses the trajectory-tracking control problem of mobile robot systems with nonholonomic constraints, in the presence of time-varying parametric uncertainties and external disturbances. This necessitates an accurate controller to meet as much control objectives, as possible. Therefore, this paper deals with an artificial intelligence control technique to design a robust controller to meet the control objectives. The design of the intelligent controller is based on the optimal control theory, the adaptive neural network system, and the robust control technique. The trajectory-tracking problem is solved using the optimal control methodology. Since the nonholonomic wheeled mobile robot is strongly nonlinear, the neural network system is applied to approximate the nonlinear function in the optimal control law. The robust controller, for his part, is then applied to adaptively estimate an unknown upper bound of the time-varying parametric uncertainties, external disturbances and approximation error of the neural network system. The stability of the closed-loop robot system is proven using the optimal control theory and Lyapunov stability analysis. The results of the simulation studies on three typical nonholonomic mobile robots are provided to demonstrate the effectiveness of the proposed controller. In addition, a comparative study with a recent robust adaptive controller shows that our proposed intelligent controller gives better results, in the sense that the output trajectory converges to the steady state faster with smaller tracking error.

- by Abdelkrim Boukabou
- •
- Neurocomputing
- by Jorge Gonzalez
- •
- Neural Network, Genetic Algorithm, Model Selection, Global Optimization

As potential candidates for explaining human cognition, connectionist models of sentence processing must demonstrate their ability to behave systematically, generalizing from a small training set. It has recently been shown that simple recurrent networks and, to a greater extent, echo-state networks possess some ability to generalize in artificial language learning tasks. We investigate this capacity for a recently introduced model that consists of separately trained modules: a recursive self-organizing module for learning temporal context representations and a feedforward two-layer perceptron module for next-word prediction. We show that the performance of this architecture is comparable with echo-state networks. Taken together, these results weaken the criticism of connectionist approaches, showing that various general recursive connectionist architectures share the potential of behaving systematically. r Bode´n and van Gelder [3] proposed a more fine-grained taxonomy of the levels of systematicity, but since here we only focus on weak systematicity, there is no need to introduce this taxonomy here. . His research areas include neural networks and cognitive science with focus on connectionist natural language modeling.

- by Igor Farkas
- •
- Engineering, Sentence Processing, Language Learning, Recurrent Neural Network