Spiking neural networks (SNNs) mimic their biological counterparts more closely than their predecessors and are considered the third generation of artificial neural networks. It has been proven that networks of spiking neurons have a higher computational capacity and lower power requirements than sigmoidal neural networks. This article introduces a new type of SNN that draws inspiration and incorporates concepts from neuronal assemblies in the human brain. The proposed network, termed as class-dependent neuronal activation-based SNN (CDNA-SNN), assigns each neuron learnable values known as CDNAs which indicate the neuron's average relative spiking activity in response to samples from different classes. A new learning algorithm that categorizes the neurons into different class assemblies based on their CDNAs is also presented. These neuronal assemblies are trained via a novel training method based on spike-timing-dependent plasticity (STDP) to have high activity for their associated class and low firing rate for other classes. Also, using CDNAs, a new type of STDP that controls the amount of plasticity based on the assemblies of pre- and postsynaptic neurons is proposed. The performance of CDNA-SNN is evaluated on five datasets from the University of California, Irvine (UCI) machine learning repository, as well as Modified National Institute of Standards and Technology (MNIST) and Fashion MNIST, using nested cross-validation (N-CV) for hyperparameter optimization. Our results show that CDNA-SNN significantly outperforms synaptic weight association training (SWAT) ( ) and SpikeProp ( ) on 3/5 and self-regulating evolving spiking neural (SRESN) ( ) on 2/5 UCI datasets while using the significantly lower number of trainable parameters. Furthermore, compared to other supervised, fully connected SNNs, the proposed SNN reaches the best performance for Fashion MNIST and comparable performance for MNIST and neuromorphic-MNIST (N-MNIST), also utilizing much less (1%-35%) parameters.

Abstract Involving and engaging people with intellectual disabilities on issues relating to their mental wellbeing is essential if relevant tools and solutions are to be developed. This research explores how inclusive and participatory co-design techniques and principles can be used to engage people with intellectual disabilities in designing innovations in mental wellbeing tangible technologies. In particular, individuals with intellectual disabilities participated in a co-design process via a series of workshops and focus groups to design tangible interfaces for mental wellbeing as their wellbeing challenges are often diagnostically overshadowed. The workshops helped participants explore new technologies, including sensors and feedback mechanisms that can help monitor and potentially improve mental wellbeing. The adopted co-design approach resulted in a range of effective and suitable interfaces being developed for varying ages.

Vertebrate retinas are highly-efficient in processing trivial visual tasks such as detecting moving objects, which still represent complex challenges for modern computers. In vertebrates, the detection of object motion is performed by specialised retinal cells named Object Motion Sensitive Ganglion Cells (OMS-GC). OMS-GC process continuous visual signals and generate spike patterns that are post-processed by the Visual Cortex. Our previous Hybrid Sensitive Motion Detector (HSMD) algorithm was the first hybrid algorithm to enhance Background subtraction (BS) algorithms with a customised 3-layer Spiking Neural Network (SNN) that generates OMS-GC spiking-like responses. In this work, we present a Neuromorphic Hybrid Sensitive Motion Detector (NeuroHSMD) algorithm that accelerates our HSMD algorithm using Field-Programmable Gate Arrays (FPGAs). The NeuroHSMD was compared against the HSMD algorithm, using the same 2012 Change Detection (CDnet2012) and 2014 Change Detection (CDnet2014) benchmark datasets. When tested against the CDnet2012 and CDnet2014 datasets, NeuroHSMD performs object motion detection at 720 × 480 at 28.06 Frames Per Second (fps) and 720 × 480 at 28.71 fps, respectively, with no degradation of quality. Moreover, the NeuroHSMD proposed in this article was completely implemented in Open Computer Language (OpenCL) and therefore is easily replicated in other devices such as Graphical Processing Units (GPUs) and clusters of Central Processing Units (CPUs).

Spiking Neural Networks (SNNs), An alternative to sigmoidal neural networks, include time into their operations using discrete signals called spikes. Employing spikes enables SNNs to mimic any feedforward sigmoidal neural network with lower power consumption. Recently a new type of SNN has been introduced for classification problems, known as Degree of Belonging SNN (DoB-SNN). DoB-SNN is a two-layer spiking neural network that shows significant potential as an alternative SNN architecture and learning algorithm. This paper introduces a new variant of Spike-Timing Dependent Plasticity (STDP), which is based on the assembly of neurons and expands the DoB-SNN's training algorithm for multilayer architectures. The new learning rule, known as assembly-based STDP, employs trained DoBs in each layer to train the next layer and build strong connections between neurons from the same assembly while creating inhibitory connections between neurons from different assemblies in two consecutive layers. The performance of the multilayer DoB-SNN is evaluated on five datasets from the UCI machine learning repository. Detailed comparisons on these datasets with other supervised learning algorithms show that the multilayer DoB-SNN can achieve better performance on 4/5 datasets and comparable performance on 5th when compared to multilayer algorithms that employ considerably more trainable parameters.

Mental health problems are on the rise globally and strain national health systems worldwide. Mental disorders are closely associated with fear of stigma, structural barriers such as financial burden, and lack of available services and resources which often prohibit the delivery of frequent clinical advice and monitoring. Technologies for mental well-being exhibit a range of attractive properties, which facilitate the delivery of state-of-the-art clinical monitoring. This review article provides an overview of traditional techniques followed by their technological alternatives, sensing devices, behaviour changing tools, and feedback interfaces. The challenges presented by these technologies are then discussed with data collection, privacy, and battery life being some of the key issues which need to be carefully considered for the successful deployment of mental health toolkits. Finally, the opportunities this growing research area presents are discussed including the use of portable tangible interfaces combining sensing and feedback technologies. Capitalising on the data these ubiquitous devices can record, state of the art machine learning algorithms can lead to the development of robust clinical decision support tools towards diagnosis and improvement of mental well-being delivery in real-time.

The rapidly increasing volumes of data and the need for big data analytics have emphasized the need for algorithms that can accommodate incomplete or noisy data. The concept of recurrency is an important aspect of signal processing, providing greater robustness and accuracy in many situations, such as biological signal processing. Probabilistic fuzzy neural networks (PFNN) have shown potential in dealing with uncertainties associated with both stochastic and nonstochastic noise simultaneously. Previous research work on this topic has addressed either the fuzzy-neural aspects or alternatively the probabilistic aspects, but currently a probabilistic fuzzy neural algorithm with recurrent feedback does not exist. In this article, a PFNN with a recurrent probabilistic generation module (designated PFNN-R) is proposed to enhance and extend the ability of the PFNN to accommodate noisy data. A back-propagation-based mechanism, which is used to shape the distribution of the probabilistic density function of the fuzzy membership, is also developed. The objective of the work was to develop an approach that provides an enhanced capability to accommodate various types of noisy data. We apply the algorithm to a number of benchmark problems and demonstrate through simulation results that the proposed technique incorporating recurrency advances the ability of PFNNs to model time-series data with high intensity, random noise.

Robots need to exploit high-quality information on grasped objects to interact with the physical environment. Haptic data can therefore be used for supplementing the visual modality. This paper investigates the use of Convolutional Neural Networks (CNN) and Long-Short Term Memory (LSTM) neural network architectures for object classification on Spatio-temporal tactile grasping data. Furthermore, we compared these methods using data from two different fingertip sensors (namely the BioTac SP and WTS-FT) in the same physical setup, allowing for a realistic comparison across methods and sensors for the same tactile object classification dataset. Additionally, we propose a way to create more training examples from the recorded data. The results show that the proposed method improves the maximum accuracy from 82.4 % (BioTac SP fingertips) and 90.7 % (WTS-FT fingertips) with complete time-series data to about 94 % for both sensor types.

Spiking neural networks (SNNs) as the third generation of artificial neural networks are closer to their biological counterparts than their predecessors. SNNs have a higher computational capacity and lower power requirements than networks of sigmoidal neurons. In this paper, a new spiking neural network for pattern classification referred to as Degree of Belonging SNN (DoB-SNN) is introduced. DoB-SNN is inspired by a neuronal assembly where each neuron has a degree of belonging to every class of data being process. DoB-SNN clusters the neurons during the training process using DoBs to allocate a group of neurons to each class. A new training algorithm is presented to adjust DoBs along with the network's synaptic weights, based on Spike-Timing Dependent Plasticity (STDP) and neurons' activity for training samples. The performance of DoB-SNN is evaluated on five datasets from the UCI machine learning repository. Nested Cross-Validation is employed to determine the network's hyperparameters for each dataset and thoroughly assess generalisation capability. A detailed comparison on these datasets with three other supervised learning algorithms, including SpikeProp, SWAT, and SRESN is provided. The results show that no algorithm significantly outperforms DoB-SNN, Whereas DoB-SNN has significantly better performance than others for Liver disorders dataset (>6.10%,$p < 0.01$). Accuracies obtained by DoB-SNN are significantly greater than SWAT for both Iris and Breast Cancer (>1.69%,$p < 0.001$) and significantly better than SpikeProp for Iris (1.62%,$p=0.04$). In all comparisons, DoB-SNN used the smallest network, among others. DoB-SNN therefore offers significant potential as alternative SNN architecture and learning algorithm.

The ability to quickly and accurately triage a person’s medical condition in an emergency situation or other critical scenarios could mean the difference between life and death. Endowing a robotic system with vision and tactile capabilities, similar to those of medical professionals, and thus enabling robots to assess a patient’s status in an emergency is a highly sought after characteristic in healthcare robotics. This paper presents a novel fuzzy triage system exploiting visual and tactile sensing, to equip a robot with the skills to accurately determine key vital signs in humans. There are three key signs of human health: respiratory rate, pulse rate (Beats Per Minute (BPM)) and capillary refill time. Using ground truth from a medical professional, the fuzzy triage system is trained and validated initially with informed synthetic data and then further evaluated using vital signs data collected from subjects in a pilot study. Results from this pilot study indicate that the fuzzy triage system is capable of classifying a patient’s health using the novel approaches for collecting BPM, Respiratory Rate (RR) and Capillary Refill Time (CRT) which replicate, to some extent, the approaches used by medical professionals for measuring vital signs. Furthermore, the intelligent system proved capable of determining whether a pulse was regular or arrhythmic, whether respiratory rate was regular or irregular, and determining the subject’s capillary refill time. Such results imply that this system could ultimately be used, for example, in a home assistance robot for elderly or disabled persons, or as a first responder robot. Ultimately the aim would be that these methods could be utilised by robotic systems in emergency scenarios or disaster zones.

Model optimisation is a key step in model development and traditionally this was limited to parameter tuning. However, recent developments and enhanced understanding of internal dynamics of model architectures have led to various exploration to optimise and enhance performance through model extension and development. In this paper, we extend the architecture of the Multi-recurrent Neural Network (MRN) to incorporate self-learning recurrent link ratios and periodically attentive hidden units. We contrast and show the superiority of these extensions to the standard MRN for a complex financial prediction task. The superiority is attributed to i) the ability of the self-learning recurrent link ratios to dynamically utilise data to identify optimal parameters of its memory mechanism and ii) the periodically attentive units enabling the hidden layer capture temporal features that are sensitive to different periods of time. Finally, we evaluate our extended MRNs (Self-Learning MRN (SL-MRN) and Periodically Attentive MRN (PA-MRN)), against two current state-of the-art models (Long-Short Term Memory and Support Vector Machines) for an eye state detection task. Our preliminary results demonstrate that the PA-MRN and SL-MRN outperform both state-of-the-art models. These results demonstrate that the MRN extensions are suitable models for machine learning applications and these findings would be further explored.

Accurate estimation of interactions in neuronal circuits is critical in understanding neural information processing and the neuronal dynamics of emergent networks. Transfer entropy(TE) is a model-free information theoretic measure of flow of information between two random processes. TE has recently gained much popularity due to its effectiveness in estimating effective connectivity among neurons from simulated networks. However, experimental recordings inherently lack the ground truth information of neural connectivity - making it difficult to identify true connections from spurious connections. To tackle this problem, we present a superimposition method where the estimated connectivity is superimposed onto the spatial firing density plot. The firing density plot, on itself, does not provide connectivity information but we assume that frequently firing channels are more likely to have more active interactions. A neuron's firing depends on the presynaptic inputs, and highly interconnected neurons tend to have more presynaptic inputs resulting in more frequent firing. The firing density plot is organised in a spatial layout, which corresponds substantially to the structural layout. We demonstrated that the connectivity estimated closely corresponds to the firing density plot when superimposed onto the firing density plot. To strengthen the arguement, we analyse the firing count and the connectivity inferred for each randomly sampled channel and found a positively correlated relationship.

Inflammatory Bowel Disease (IBD) is an umbrella term for a group of inflammatory diseases of the gastrointestinal tract, including Crohn's Disease and ulcerative colitis. Changes to the intestinal microbiome, the community of micro-organisms that resides in the human gut, have been shown to contribute to the pathogenesis of IBD. IBD diagnosis is often delayed due to its non-specific symptoms and because an invasive colonoscopy is required for confirmation, which leads to poor growth in children and worse treatment outcomes. Feature selection algorithms are often applied to microbial communities to identify bacterial groups that drive disease. It has been shown that aggregating Ensemble Feature Selection (EFS) can improve the robustness of feature selection algorithms, which is defined as the variation of feature selector output caused by small changes to the dataset. In this work, we apply a two-step filter and an EFS process to generate robust feature subsets that can non-invasively predict IBD subtypes from high-resolution microbiome data. The predictive power of the robust feature subsets is the highest reported in literature to date. Furthermore, we identify five biologically plausible bacterial species that have not previously been implicated in IBD aetiology.

Spike train synchrony estimation of neuronal cultures provides valuable insights into firing patterns of neurons in terms of degree of similarity or dissimilarity. These estimations have proven to be a useful tool in neuroscience since synchrony in neuronal networks is thought to be related to cognitive processes, sensory awareness, learning and neurological disorders. Many mathematical measures have been developed to quantify the degree of synchrony. These synchrony metrics are generally used for smaller sets of spike trains and not been explored for larger High Density Multi Electrode Arrays (HD-MEA) datasets with thousands of channels. Here, bivariate and multivariate ISI-distance and SPIKE-distance metrics are utilized on both synthetic and experimental HD-MEA datasets to quantify spike train synchrony. It is demonstrated that, despite the significant size of the datasets, the approaches are effective in identifying and quantifying interesting bursting or change in spike train behaviours which are not always obvious from the raster plot.

The retina performs the earlier stages of image processing in living beings and is composed of six different groups of cells, namely, the rods, cones, horizontal, bipolar, amacrine and ganglion cells. Each of those group of cells can be sub-divided into other types of cells that vary in shape, size, connectivity and functionality. Each cell is responsible for performing specific tasks in these early stages of biological image processing. Some of those cells are sensitive to horizontal and vertical movements. This paper proposes a multi-hierarchical spiking neural network architecture for detecting horizontal and vertical movements using a custom dataset which was generated in laboratory settings. The proposed architecture was designed to reflect the connectivity, behaviour and the number of layers found in the majority of vertebrates retinas, including humans. The architecture was trained using 2303 images and tested using 816 images. Simulation results revealed that each cell model is sensitive to vertical and horizontal movements with a detection error of 6.75 percent.

Determining the receptive field of a retinal ganglion cell is critically important when formulating a computational model that maps the relationship between the stimulus and response. This process is traditionally undertaken using reverse correlation to estimate the receptive field. By stimulating the retina with artificial stimuli, such as alternating checkerboards, bars or gratings and recording the neural response it is possible to estimate the cell’s receptive field by analysing the stimuli that produced the response. Artificial stimuli such as white noise is known to not stimulate the full range of the cell’s responses. By using natural image stimuli, it is possible to estimate the receptive field and obtain a resulting model that more accurately mimics the cells’ responses to natural stimuli. This paper extends on previous work to seek further improvements in estimating a ganglion cell’s receptive field by considering that the receptive field can be divided into subunits. It is thought that these subunits may relate to receptive fields which are associated with bipolar retinal cells. The findings of this preliminary study show that by using subunits to define the receptive field we achieve a significant improvement over existing approaches when deriving computational models of the cell’s response.

For a considerable time, it has been the goal of computational neuroscientists to understand biological nervous systems. However, the vast complexity of such systems has made it very difficult to fully understand even basic functions such as movement. Because of its small neuron count, the C. elegans nematode offers the opportunity to study a fully described connectome and attempt to link neural network activity to behaviour. In this paper a simulation of the neural network in C. elegans that responds to chemical stimulus is presented and a consequent realistic head movement demonstrated. An evolutionary algorithm (EA) has been utilised to search for estimates of the values of the synaptic conductances and also to determine whether each synapse is excitatory or inhibitory in nature. The chemotaxis neural network was designed and implemented, using the parameterisation obtained with the EA, on the Si elegans platform a state-of-the-art hardware emulation platform specially designed to emulate the C. elegans nematode.

A Deep Boltzmann Machine is a model of a Deep Neural Network formed from multiple layers of neurons with nonlinear activation functions. The structure of a Deep Boltzmann Machine enables it to learn very complex relationships between features and facilitates advanced performance in learning of high-level representation of features, compared to conventional Artificial Neural Networks. Feature selection at the input level of Deep Neural Networks has not been well studied, despite its importance in reducing the input features processed by the deep learning model, which facilitates understanding of the data. This paper proposes a novel algorithm, Deep Feature Selection (Deep-FS), which is capable of removing irrelevant features from large datasets in order to reduce the number of inputs which are modelled during the learning process. The proposed Deep-FS algorithm utilizes a Deep Boltzmann Machine, and uses knowledge which is acquired during training to remove features at the beginning of the learning process. Reducing inputs is important because it prevents the network from learning the associations between the irrelevant features which negatively impact on the acquired knowledge of the network about the overall distribution of the data. The Deep-FS method embeds feature selection in a Restricted Boltzmann Machine which is used for training a Deep Boltzmann Machine. The generative property of the Restricted Boltzmann Machine is used to reconstruct eliminated features and calculate reconstructed errors, in order to evaluate the impact of eliminating features. The performance of the proposed approach was evaluated with experiments conducted using the MNIST, MIR-Flickr, GISETTE, MADELON and PANCAN datasets. The results revealed that the proposed Deep-FS method enables improved feature selection without loss of accuracy on the MIR-Flickr dataset, where Deep-FS reduced the number of input features by removing 775 features without reduction in performance. With regards to the MNIST dataset, Deep-FS reduced the number of input features by more than 45%; it reduced the network error from 0.97% to 0.90%, and also reduced processing and classification time by more than 5.5%. Additionally, when compared to classical feature selection methods, Deep-FS returned higher accuracy. The experimental results on GISETTE, MADELON and PANCAN showed that Deep-FS reduced 81%, 57% and 77% of the number of input features, respectively. Moreover, the proposed feature selection method reduced the classifier training time by 82%, 70% and 85% on GISETTE, MADELON and PANCAN datasets, respectively. Experiments with various datasets, comprising a large number of features and samples, revealed that the proposed Deep-FS algorithm overcomes the main limitations of classical feature selection algorithms. More specifically, most classical methods require, as a prerequisite, a pre-specified number of features to retain, however in Deep-FS this number is identified automatically. Deep-FS performs the feature selection task faster than classical feature selection algorithms which makes it suitable for deep learning tasks. In addition, Deep-FS is suitable for finding features in large and big datasets which are normally stored in data batches for faster and more efficient processing.

In recent years, artificial vision research has moved from focusing on the use of only intensity images to include using depth images, or RGB-D combinations due to the recent development of low-cost depth cameras. However, depth images require a lot of storage and processing requirements. In addition, it is challenging to extract relevant features from depth images in real time. Researchers have sought inspiration from biology in order to overcome these challenges resulting in biologically inspired feature extraction methods. By taking inspiration from nature, it may be possible to reduce redundancy, extract relevant features, and process an image efficiently by emulating biological visual processes. In this paper, we present a depth and intensity image feature extraction approach that has been inspired by biological vision systems. Through the use of biologically inspired spiking neural networks, we emulate functional computational aspects of biological visual systems. The results demonstrate that the proposed bioinspired artificial vision system has increased performance over existing computer vision feature extraction approaches.

The processing capabilities of biological vision systems are still vastly superior to artificial vision, even though this has been an active area of research for over half a century. Current artificial vision techniques integrate many insights from biology yet they remain far-off the capabilities of animals and humans in terms of speed, power, and performance. A key aspect to modeling the human visual system is the ability to accurately model the behavior and computation within the retina. In particular, we focus on modeling the retinal ganglion cells (RGCs) as they convey the accumulated data of real world images as action potentials onto the visual cortex via the optic nerve. Computational models that approximate the processing that occurs within RGCs can be derived by quantitatively fitting the sets of physiological data using an input-output analysis where the input is a known stimulus and the output is neuronal recordings. Currently, these input-output responses are modeled using computational combinations of linear and nonlinear models that are generally complex and lack any relevance to the underlying biophysics. In this paper, we illustrate how system identification techniques, which take inspiration from biological systems, can accurately model retinal ganglion cell behavior, and are a viable alternative to traditional linear-nonlinear approaches.

A fundamental problem in cooperative HumanRobot Interaction is object handover. Existing works in this area assume the human can reliably grasp the object from the robot hand. However, in some situations the human can produce perturbing forces in the object that are not meant to end in a handover. These perturbations can result in the object being dropped or the robot hand being damaged. This paper addresses this problem and presents a mechanism for reliable robot to human object handover implemented in a Shadow Robot hand endowed with tactile sensing. Given a stable grasping configuration, using BioTAC sensors we are able to estimate the contact forces applied to the object, and provide a feedback signal to a joint effort controller to maintain grasp forces despite perturbations. Our system is able to identify between object pulling forces which should result in an object handover, and other disturbances. Experimental results show that the hand releases the object only when the object is pulled, validating the proposed algorithm.

A fundamental aspect of many human-robot collaborative tasks is object exchange or handover. Several techniques have been proposed to decide when a robot hand or gripper should release an object for a human to receive. However, these techniques typically neglect the reliability of the handover, assuming the process will occur without issue. Based on the fact that humans apply pulling forces in specific directions when receiving an object, this paper presents a recursive procedure enabling a robot to release an object appropriately and timely during a handover. Experiments with naive users showed subject-specific consistent pulling directions during robot-human handovers, highlighting the need for a system to be capable of detecting force directions in relation to objects. The approach reported in this paper shows that through tactile sensing the proposed approach can accurately classify between five different actions impacting an object held by a robot hand. The recursive nature of the system also enables detection of sequences of different actions, enabling the robot to decide to safely release the object only when the pulling performed by the human is in the right direction.

The creation of a predictive system that correctly forecasts future changes of a stock price is crucial for investment management and algorithmic trading. The use of technical analysis for financial forecasting has been successfully employed by many researchers. Input window length is a time frame parameter required to be set when calculating many technical indicators. This study explores how the performance of the predictive system depends on a combination of a forecast horizon and an input window length for forecasting variable horizons. Technical indicators are used as input features for machine learning algorithms to forecast future directions of stock price movements. The dataset consists of ten years daily price time series for fifty stocks. The highest prediction performance is observed when the input window length is approximately equal to the forecast horizon. This novel pattern is studied using multiple performance metrics: prediction accuracy, winning rate, return per trade and Sharpe ratio.

One of the urgent challenges in the automated analysis and interpretation of electrical brain activity is the effective handling of uncertainties associated with the complexity and variability of brain dynamics, reflected in the nonstationary nature of brain signals such as electroencephalogram (EEG). This poses a severe problem for existing approaches to the classification task within brain-computer interface (BCI) systems. Recently emerged type-2 fuzzy logic (T2FL) methodology has shown a remarkable potential in dealing with uncertain information given limited insight into the nature of the data-generating mechanism. The objective of this work is, thus, to examine the applicability of the T2FL approach to the problem of EEG pattern recognition. In particular, the focus is two-fold: 1) the design methodology for the interval T2FL system (IT2FLS) that can robustly deal with intersession as well as within-session manifestations of nonstationary spectral EEG correlates of motor imagery, and 2) the comprehensive examination of the proposed fuzzy classifier in both off-line and on-line EEG classification case studies. The on-line evaluation of the IT2FLS-controlled real-time neurofeedback over multiple recording sessions holds special importance for EEG-based BCI technology. In addition, a retrospective comparative analysis accounting for other popular BCI classifiers such as linear discriminant analysis, kernel Fisher discriminant, and support vector machines as well as a conventional type-1 FLS, simulated off-line on the recorded EEGs, has demonstrated the enhanced potential of the proposed IT2FLS approach to robustly handle uncertainty effects in BCI classification.

This paper presents an enhanced rank-order-based learning algorithm, called SpikeTemp, for spiking neural networks (SNNs) with a dynamically adaptive structure. The trained feed-forward SNN consists of two layers of spiking neurons: 1) an encoding layer which temporally encodes real-valued features into spatio-temporal spike patterns and 2) an output layer of dynamically grown neurons which perform spatio-temporal classification. Both Gaussian receptive fields and square cosine population encoding schemes are employed to encode real-valued features into spatio-temporal spike patterns. Unlike the rank-order-based learning approach, SpikeTemp uses the precise times of the incoming spikes for adjusting the synaptic weights such that early spikes result in a large weight change and late spikes lead to a smaller weight change. This removes the need to rank all the incoming spikes and, thus, reduces the computational cost of SpikeTemp. The proposed SpikeTemp algorithm is demonstrated on several benchmark data sets and on an image recognition task. The results show that SpikeTemp can achieve better classification performance and is much faster than the existing rank-order-based learning approach. In addition, the number of output neurons is much smaller when the square cosine encoding scheme is employed. Furthermore, SpikeTemp is benchmarked against a selection of existing machine learning algorithms, and the results demonstrate the ability of SpikeTemp to classify different data sets after just one presentation of the training samples with comparable classification performance.

A system of interacting agents is, by definition, very demanding in terms of computational resources. Although multi-agent systems have been used to solve complex problems in many areas, it is usually very difficult to perform large-scale simulations in their targeted serial computing platforms. Reconfigurable hardware, in particular Field Programmable Gate Arrays devices, have been successfully used in High Performance Computing applications due to their inherent flexibility, data parallelism, and algorithm acceleration capabilities. Indeed, reconfigurable hardware seems to be the next logical step in the agency paradigm, but only a few attempts have been successful in implementing multi-agent systems in these platforms. This article discusses the problem of inter-agent communications in Field Programmable Gate Arrays. It proposes a Network-on-Chip in a hierarchical star topology to enable agents’ transactions through message broadcasting using the Open Core Protocol as an interface between hardware modules. A customizable router microarchitecture is described and a multi-agent system is created to simulate and analyse message exchanges in a generic heavy traffic load agent-based application. Experiments have shown a throughput of 1.6Gbps per port at 100MHz without packet loss and seamless scalability characteristics.

The emergence of social media has provided vast amounts of information that is potentially valuable for emergency management. In the EU-FP7 Project Security Systems for Language and Image Analysis (Slandail), an image analysis system has been developed to recognize the flood water images from the social media resources by incorporating with text analysis. A novel image feature descriptor has been developed to facilitate fast image processing based on incorporation of the "Squiral" (Square-Spiral) Image Processing (SIP) framework with the "Speeded-up Robust Features" (SURF). A new approach is proposed to generate an index from image recognition outcomes based on a moving window average, which presents a temporal change based on the occurrence of flooding water identified by image analysis. The evaluation for computation time and recognition were based on a batch of images obtained from the US Federal Emergency Management Agency (FEMA) media library and Facebook corpus from Germany, and the outcomes show the advantages of the proposed image features. The simulation results demonstrate the concept of the index based on a moving window average, highlighting the potential for application in emergency management.

The nematode C. elegans (Caenorhabditis elegans) has for many years been instrumental as a model organism for fundamental research into biological neural networks, mainly to understand the behaviour and physiology of nervous systems. The Si elegans EU FP7 project aims to develop a Hardware Neural Network (HNN) to accurately replicate the C. elegans nervous system, behaviour and response to environmental changes that will enable neuroscientists to better understand these basic functions. In this paper, we focus on consideration of some specific behaviours of the worm for modelling in the hardware based Si elegans system. Environmental response to oxygen, food, light, temperature and other chemicals are reviewed to establish specific neurons responsible for the worm behaviours. Neuron interconnects mechanisms and suitable models simulating the biological worm are also discussed.

Tactile sensing has recently attracted significant research interest in robotics. Despite the fact that tactile sensors provide temporal sequences of readings, state-of-the-art material recognition approaches are episodic, i.e. a whole sequence of readings is processed to identify the material. Based on vibration frequency response, this work presents an online identification technique using recursive estimation of the probability of identifying a set of materials, i.e. casting the classification problem as a hidden Markov model (HMM) state estimation problem. This allows for faster identification of most materials and does not require several exploratory movements. Our results show that when enough evidence is gathered, the system eventually achieves perfect recognition of our experimental set of 34 materials with an average identification time of ≤ 0.5 seconds. To prove the accuracy of this method, we also conducted a comparative experiment with commonly used machine learning algorithms for material identification such as k-Nearest Neighbour(KNN), an Artificial Neural Network(ANN) and Support Vector Machine(SVM).

Tinnitus is the phantom perception of a sound heard in or around the head in the absence of an identifiable source affecting 10-15% worldwide. The majority of tinnitus sufferers have some form of hearing loss. The multiple pathologies that generate and sustain tinnitus in a diverse tinnitus population make it challenging to establish a homogeneous cohort for experimental studies. People with no hearing loss or previous experience of tinnitus also begin to perceive phantom sounds when situated in a sound proof room for five minutes or less. This is consistently observed across multiple studies. Studies that induced tinnitus through acoustic deprivation in healthy subjects provide a more controlled environment to observe tinnitus. Although experimental work shows what is happening it does not explain how the tinnitus related activity is generated. Computational modelling of tinnitus following hearing loss has shown that underlying mechanisms, such as adaptive gain, can generate hyperactivity in the regions of hearing loss. These models do not account for the generation of tinnitus in people with no hearing loss. In this work we model the development of tinnitus related activity in cases of no hearing loss and induced acoustic deprivation. The tinnitus related activity disappears once the model is returned to normal ambient noise.

Understanding how simple biological entities perform complex tasks (such as moving away from predators, reproduction and seeking for food) is one of the main goals of researchers across the world. The C. elegans nematode is one of the simplest and well characterised biological nervous systems (BNS) and the Si elegans EU FP7 project aims to develop a framework to mimic its functionality and better understand basic functions. This paper presents a small scale emulation environment composed of 8 neurons running on field programmable gate arrays (FPGAs), where each neuron FPGA is interconnected to all of the other neuron FPGAs via an innovative optical synaptic connectivity strategy. This paper evaluates the fidelity of synaptic communication for transmission of spikes using both wired and optical methods.

The nematode C. elegans is the only organism to have its connectome fully documented. All 302 neurons of the worm are already known, however, their description is somewhat lacking. Using current computational techniques for clustering, such as Growing Self Organizing Networks and K-Means algorithm, this paper aims to provide a deeper analysis and understanding of the C. elegans connectome and the particulars of its individual neurons. The algorithms show that there are several groups of neurons, although the rational for the grouping is not clear and the clusters are inconsistent.

Inflammatory bowel disease (IBD) is a group of inflammatory diseases of the human colon and small intestine. IBD symptoms are non-specific; diagnosis can be delayed because an invasive colonoscopy is required for confirmation. Delayed diagnosis is linked to poor growth in children. Imbalances in the human intestinal microbiome - the community of microorganisms that reside in the human gut - are thought to contribute to the development of IBD. Work done to date in classifying host health statuses from patterns in human microbiomes with supervised learning algorithms has focused on modelling what is present in the gut (i.e. a bacterial census) with the random forest algorithm. Metagenomic shotgun sequencing is required to understand what is occurring in the gut (i.e. gene functions) and is often cost prohibitive for hundreds of samples. However, gene functions can be predicted with the Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PiCRUSt) software package, which could represent a valuable source of new features. In this paper we investigate feature relevance across the feature set with the Boruta algorithm. We find that the majority of relevant features are from the predicted metagenome. Support vector machines (SVM) and multilayer perceptrons (MLP) are rarely used with microbiomic datasets but offer several theoretical advantages. To determine if the new features and alternative algorithms are appropriate, we experiment with a range of machine learning and computational intelligence algorithms. With the best performing algorithms we also implement a conditional multiple classifier system that can identify IBD presence, IBD subtype, and IBD activity from a non-invasive stool sample.

Tactile sensing has been used in robotics for object identification, grasping, and material recognition. Most material recognition approaches use vibration signals from a tactile exploration, typically above one second long, to identify the material. This work proposes a tactile multi-modal (vibration and thermal) material identification approach based on recursive Bayesian estimation. Through the frequency response of the vibration induced by the material and thermal features, like an estimate of the thermal power loss of the finger, we show that it is possible to identify materials in less than half a second. Moreover, a comparison between vibration only and multi-modal identification shows that both recognition time and classification errors are reduced by adding thermal information.

A wash trade refers to the illegal activities of traders who utilize carefully designed limit orders to manually increase the trading volumes for creating a false impression of an active market. As one of the primary formats of market abuse, a wash trade can be extremely damaging to the proper functioning and integrity of capital markets. The existing work focuses on collusive clique detections based on certain assumptions of trading behaviors. Effective approaches for analyzing and detecting wash trade in a real-life market have yet to be developed. This paper analyzes and conceptualizes the basic structures of the trading collusion in a wash trade by using a directed graph of traders. A novel method is then proposed to detect the potential wash trade activities involved in a financial instrument by first recognizing the suspiciously matched orders and then further identifying the collusions among the traders who submit such orders. Both steps are formulated as a simplified form of the knapsack problem, which can be solved by dynamic programming approaches. The proposed approach is evaluated on seven stock data sets from the NASDAQ and the London Stock Exchange. The experimental results show that the proposed approach can effectively detect all primary wash trade scenarios across the selected data sets.

Fast image processing is a key element in achieving real-time image and video analysis. The spiral addressing scheme [10] has been an efficient tool for hexagonal image processing (HIP), whereby the image pixel indices are stored in a one-dimensional vector that enables fast processing. Unlike HIP, which requires a complex resampling scheme, we present a novel “squiral” (square spiral) image processing (SIP) framework that provides a spiral addressing scheme for direct application to standard square pixel-based images. A SIP-based non-overlapping convolution technique is developed by simulating the eye tremor phenomenon of the human visual system to accelerate computation in feature extraction. Furthermore, we deploy the proposed simulated eye tremor technique on a sequence of video frames. The preliminary results based on two action video clips demonstrate the potential of the SIP-based eye tremor model to facilitate fast video processing.

Pulse rate and rhythm are indicators of the health of a human's blood circulation. Being able to detect one's pulse rate and rhythm in an emergency situation could be the difference between life and death. The work presented in this paper is preliminary work on algorithms that will equip a robot with the necessary skills to assess a human's pulse. Algorithms for pulse detection and the calculation of beats per minute (bpm) and Pulse to Pulse Interval (PPI) are presented and used to classify if a heart rate is normal, bradycardic, or tachycardic. Furthermore PPI is used to determine if the pulse rate is regular or in a form of arrhythmia. The results in this paper show that pulse was successfully detected and a subjects bpm calculated using a trough detection method. Furthermore the system proved to be capable of classifying the pulse as being regular or arrhythmic.

The hardware layer of the Si elegans EU FP7 project is a massively parallel architecture designed to accurately emulate the C. elegans nematode in biological real-time. The C. elegans nematode is one of the simplest and well characterized Biological Nervous Systems (BNS) yet many questions related to basic functions such as movement and learning remain unanswered. The hardware layer includes a Hardware Neural Network (HNN) composed of 302 FPGAs (one per neuron), a Hardware Muscle Network (HMN) composed of 27 FPGAs (one per 5 muscles) and one Interface Manager FPGA, which is physically connected through 2 Local Area Networks (LANs) and through an innovative 3D optical connectome. Neuron structures (gap junctions and synapses) and muscles are modelled in the design environment of the software layer and their simulation data (spikes, variable values and parameters) generate data packets sent across the Local Area Networks (LAN). Furthermore, a software layer gives the user a set of design tools giving the required flexibility and high level hardware abstraction to design custom neuronal models. In this paper the authors present an overview of the hardware layer, connections infrastructure and communication protocol.

Even though artificial vision has been in development for over half a century it still fares poorly when compared to biological vision. The processing capabilities of biological visual systems are vastly superior in terms of power, speed, and performance. Inspired by this robust performance artificial vision systems have sought to take inspiration from biology by modeling aspects of biological vision systems. Existing computational models of visual neurons can be derived by quantitatively fitting particular sets of physiological data using an input-output analysis where a known input is given to the system and its output is recorded. These models need to capture the full spatio-temporal description of neuron behaviour under natural viewing conditions. In this work we use state-of-the-art fuzzy neural network techniques to accurately model the responses of retinal ganglion cells. We illustrate how a self-organising fuzzy neural network can accurately model ganglion cell behaviour, and are a viable alternative to traditional system identification techniques.

Accurate forecasting of upcoming trends in the capital markets is extremely important for algorithmic trading and investment management. Before making a trading decision, investors estimate the probability that a certain news item will influence the market based on the available information. Speculation among traders is often caused by the release of a breaking news article and results in price movements. Publications of news articles influence the market state that makes them a powerful source of data in financial forecasting. Recently, researchers have developed trend and price prediction models based on information extracted from news articles. However, to date no previous research that investigates the advantages of using news articles with different levels of relevance to the target stock has been conducted. This research study uses the multiple kernel learning technique to effectively combine information extracted from stock-specific and sub-industry-specific news articles for prediction of an upcoming price movement. News articles are divided into these two categories based on their relevance to a targeted stock and analyzed by separate kernels. The experimental results show that utilizing two categories of news improves the prediction accuracy in comparison with methods based on a single news category.

Modelling aspects of the human vision system, including the retina, is difficult due to insufficient knowledge about the internal components, organisation and complexity of the interactions within the system. Retinal ganglion cells are considered a core component of the human visual system as they convey the accumulated data as action potentials onto the optic nerve. Current techniques capable of mapping this input-output response involve computational combinations of linear and nonlinear models that are generally complex and lack any relevance to the underlying biophysics. This paper aims to model a retinal ganglion cell with a simple spiking neuron combined with a pre-processing method, which accounts for the preceding retinal neural structure. Performance of the models is compared with the spike responses obtained in the electrophysiological recordings from a mammalian retina subjected to visual stimulation.

This paper presents a novel RBF-like fast dynamically Evolving Spiking Neural classifier (ESNC). The trained feed-forward SNN consists of three layers of spiking neurons: an encoding layer which temporally encodes real valued features into spatio-temporal spike patterns, a hidden layer of dynamically grown and pruned neurons which perform spatiotemporal clustering, and an evolving output layer for classification. Both the structure and weights of the SNN are learned dynamically through a combination of unsupervised and supervised learning paradigms. An unsupervised clustering method is implemented by the hidden layer for adjusting the synaptic weights of the hidden neurons afferent connections. The centre of each hidden RBF neuron is represented by a vector of temporal distances between the first spike of the hidden neuron and the presynaptic spikes. In addition, strategies are proposed to adjust the structure of the hidden and output layers as inputs are presented to the SNN, and classification at the output layer is achieved through supervised learning where a learning window is used to adjust the weights of the output neurons afferent connections. The proposed learning algorithm is demonstrated on several benchmark datasets from the UCL machine learning repository. The results show comparable performance with existing machine learning algorithms and demonstrate the ability of the proposed algorithm to learn incoming data samples in a hybrid way and in one epoch only.

Despite the emerging applications of diffusion tensor imaging (DTI) to mild traumatic brain injury (mTBI), very few investigations have been reported related to temporal changes in quantitative diffusion patterns, which may help to assess recovery from head injury and the long term impact associated with cognitive and behavioral impairments caused by mTBI. Most existing methods are focused on detection of mTBI affected regions rather than quantification of temporal changes following head injury. Furthermore, most methods rely on large data samples as required for statistical analysis and, thus, are less suitable for individual case studies. In this paper, we introduce an approach based on spatial group independent component analysis (GICA), in which the diffusion scalar maps from an individual mTBI subject and the average of a group of controls are arranged according to their data collection time points. In addition, we propose a constrained GICA (CGICA) model by introducing the prior information into the GICA decomposition process, thus taking available knowledge of mTBI into account. The proposed method is evaluated based on DTI data collected from American football players including eight controls and three mTBI subjects (at three time points post injury). The results show that common spatial patterns within the diffusion maps were extracted as spatially independent components (ICs) by GICA. The temporal change of diffusion patterns during recovery is revealed by the time course of the selected IC. The results also demonstrate that the temporal change can be further influenced by incorporating the prior knowledge of mTBI (if available) based on the proposed CGICA model. Although a small sample of mTBI subjects is studied, as a proof of concept, the preliminary results provide promising insight for applications of DTI to study recovery from mTBI and may have potential for individual case studies in practice.

Price manipulation refers to the activities of those traders who use carefully designed trading behaviors to manually push up or down the underlying equity prices for making profits. With increasing volumes and frequency of trading, price manipulation can be extremely damaging to the proper functioning and integrity of capital markets. The existing literature focuses on either empirical studies of market abuse cases or analysis of particular manipulation types based on certain assumptions. Effective approaches for analyzing and detecting price manipulation in real time are yet to be developed. This paper proposes a novel approach, called adaptive hidden Markov model with anomaly states (AHMMAS) for modeling and detecting price manipulation activities. Together with wavelet transformations and gradients as the feature extraction methods, the AHMMAS model caters to price manipulation detection and basic manipulation type recognition. The evaluation experiments conducted on seven stock tick data from NASDAQ and the London Stock Exchange and 10 simulated stock prices by stochastic differential equation show that the proposed AHMMAS model can effectively detect price manipulation patterns and outperforms the selected benchmark models.

Market abuse has attracted much attention from financial regulators around the world but it is difficult to fully prevent. One of the reasons is the lack of thoroughly studies of the market abuse strategies and the corresponding effective market abuse approaches. In this paper, the strategies of reported price manipulation cases are analysed as well as the related empirical studies. A transformation is then defined to convert the time-varying financial trading data into pseudo-stationary time series, where machine learning algorithms can be easily applied to the detection of the price manipulation. The evaluation experiments conducted on four stocks from NASDAQ show a promising improved performance for effectively detecting such manipulation cases.

Wash trade refers to the activities of traders who utilise deliberately designed collusive transactions to increase the trading volumes for creating active market impression. Wash trade can be damaging to the proper functioning and integrity of capital markets. Existing work focuses on collusive clique detections based on certain assumptions of trading behaviours. Effective approaches for analysing and detecting wash trade in a real-life market have yet to be developed. This paper proposes a new analysis approach for abstracting the basic structures of wash trade based on the network topology theory and a novel approach for detecting wash trade activities. The evaluation experiments conducted on four NASDAQ stocks suggest that wash trade actions can be effectively identified based on the proposed algorithm.

Electronic trading in global markets and exchanges requires sophisticated communication and data management systems. Novel computational infrastructures and trading strategies are required to support the massive amount of incoming streaming data, where the main problem is in latency management. Multi-agent Systems have been recognized as a promising solution to address complex problems in many areas such as biology, social sciences and financial markets and may provide powerful and flexible solutions for implementing trading engines. In addition, reconfigurable hardware based on Field Programmable Gate Arrays (FPGAs) offers many important performance benefits over software implementations, such as reducing decision making latency and high-throughput data processing. Robust and scalable trading engines can be developed by leveraging the benefits of reconfigurable FPGA platforms. This paper presents a multi-agent architecture in reconfigurable hardware for financial applications and the implementation of a trading engine for pre-trade analysis as a validation scenario. Performance results show that calculation of technical indicators and trading strategy evaluation to generate trading signals with a latency of 550 ns is achievable.

In recent years, machine learning algorithms have become increasingly popular in financial forecasting. Their flexible, data-driven nature makes them ideal candidates for dealing with complex financial data. This paper investigates the effectiveness of a number of machine learning algorithms, and combinations of these algorithms, at generating one-step ahead forecasts of a number of financial time series. We find that hybrid models consisting of a linear statistical model and a nonlinear machine learning algorithm are effective at forecasting future values of the series, particularly in terms of the future direction of the series.

Accurate forecasting of directional changes in stock prices is important for algorithmic trading and investment management. Technical analysis has been successfully used in financial forecasting and recently researchers have explored the optimization of parameters for technical indicators. This study investigates the relationship between the window size used for calculating technical indicators and the accuracy of one-step-ahead (variable steps) forecasting. The directions of the future price movements are predicted using technical analysis and machine learning algorithms. Results show a correlation between window size and forecasting step size for the Support Vector Machines approach but not for the other approaches.

Online financial textual information contains a large amount of investor sentiment, i.e. subjective assessment and discussion with respect to financial instruments. An effective solution to automate the sentiment analysis of such large amounts of online financial texts would be extremely beneficial. This paper presents a natural language processing (NLP) based pre-processing approach both for noise removal from raw online financial texts and for organizing such texts into an enhanced format that is more usable for feature extraction. The proposed approach integrates six NLP processing steps, including a developed syntactic and semantic combined negation handling algorithm, to reduce noise in the online informal text. Three-class sentiment classification is also introduced in each system implementation. Experimental results show that the proposed pre-processing approach outperforms other pre-processing methods. The combined negation handling algorithm is also evaluated against three standard negation handling approaches.

Effective robotic grasping and manipulation requires knowledge about the surface properties of an object and the environment in which it is located. Physical contact with materials using tactile sensors can enable the retrieval of detailed information about the material, i.e. compressibility, surface texture and thermal properties. This paper describes a system used to classify a wide range of materials based on their thermal properties and surface texture. Following acquisition of data from a sophisticated tactile sensor, the system uses principal component analysis (PCA) to extract features from the data which are used to train an Artificial Neural Network (ANN) to classify materials, first into groups and then as individual materials. The system is compared with human performance and the results demonstrate that the proposed system performed better than humans by almost 10%.

A novel neural information processing architecture inspired by quantum mechanics and incorporating the well-known Schrodinger wave equation is proposed in this paper. The proposed architecture referred to as recurrent quantum neural network (RQNN) can characterize a nonstationary stochastic signal as time-varying wave packets. A robust unsupervised learning algorithm enables the RQNN to effectively capture the statistical behavior of the input signal and facilitates the estimation of signal embedded in noise with unknown characteristics. The results from a number of benchmark tests show that simple signals such as dc, staircase dc, and sinusoidal signals embedded within high noise can be accurately filtered and particle swarm optimization can be employed to select model parameters. The RQNN filtering procedure is applied in a two-class motor imagery-based brain-computer interface where the objective was to filter electroencephalogram (EEG) signals before feature extraction and classification to increase signal separability. A two-step inner-outer fivefold cross-validation approach is utilized to select the algorithm parameters subject-specifically for nine subjects. It is shown that the subject-specific RQNN EEG filtering significantly improves brain-computer interface performance compared to using only the raw EEG or Savitzky-Golay filtered EEG across multiple sessions.

Machine learning enables the creation of a nonlinear mapping that describes robot-environment interaction, whereas computing linguistics make the interaction transparent. In this paper, we develop a novel application of a linguistic decision tree for a robot route learning problem by dynamically deciding the robot's behavior, which is decomposed into atomic actions in the context of a specified task. We examine the real-time performance of training and control of a linguistic decision tree, and explore the possibility of training a machine learning model in an adaptive system without dual CPUs for parallelization of training and control. A quantified evaluation approach is proposed, and a score is defined for the evaluation of a model's robustness regarding the quality of training data. Compared with the nonlinear system identification nonlinear auto-regressive moving average with eXogeneous inputs model structure with offline parameter estimation, the linguistic decision tree model with online linguistic ID3 learning achieves much better performance, robustness, and reliability.

A major challenge in two-class brain-computer interface (BCI) systems is the low bandwidth of the communication channel, especially while communicating and controlling assistive devices, such as a smart wheelchair or a telepresence mobile robot, which requires multiple motion command options in the form of forward, left, right, backward, and start/stop. To address this, an adaptive user-centric graphical user interface referred to as the intelligent adaptive user interface (iAUI) based on an adaptive shared control mechanism is proposed. The iAUI offers multiple degrees-of-freedom control of a robotic device by providing a continuously updated prioritized list of all the options for selection to the BCI user, thereby improving the information transfer rate. Results have been verified with multiple participants controlling a simulated as well as physical pioneer robot.

A self-organising fuzzy-neural network (SOFNN) adapts its structure based on variations of the input data. Conventionally in such self-organising networks, the number of inputs providing the data is fixed. In this paper, we consider the situation where the number of inputs to a network changes dynamically during its online operation. We extend our existing work on a SOFNN such that the SOFNN can self-organise its structure based not only on its input data, but also according to the changes in the number of its inputs. We apply the approach to a smart home application, where there are certain situations when some of the existing events may be removed or new events emerge, and illustrate that our approach enhances cognitive reasoningin a dynamic smart home environment. In this case, the network identifies the removed and/or added events from the received information over time, and reconfigures its structure dynamically. We present results for different combinations of training and testing phases of the dynamic reconfigurable SOFNN using a set of realistic synthesized data. The results show the potential of the proposed method.

A filtering methodology inspired by the principles of quantum mechanics and incorporating the well-known Schrodinger wave equation is investigated for the first time for filtering EMG signals. This architecture, referred to as a Recurrent Quantum Neural Network (RQNN) can characterize a non-stationary stochastic signal as time varying wave packets. An unsupervised learning rule allows the RQNN to capture the statistical behaviour of the input signal and facilitates estimation of an EMG signal embedded in noise with unknown characteristics. Results from a number of benchmark tests show that simple signals such as DC, staircase DC and sinusoidal signals embedded with a high level of noise can be accurately filtered. Particle swarm optimization is employed to select RQNN model parameters for filtering simple signals. In this paper, we present the RQNN filtering procedure, using heuristically selected parameters, to be applied to a new thirteen class EMG based finger movement detection system, for emulation in a Shadow Robotics robot hand. It is shown that the RQNN EMG filtering improves the classification performance compared to using only the raw EMG signals, across multiple feature extraction approaches and subjects. Effective control of the robot hand is demonstrated.

The recent development of low cost cameras that capture 3-dimensional images has changed the focus of computer vision research from using solely intensity images to the use of range images, or combinations of RGB, intensity and range images. The low cost and widespread availability of the hardware to capture these images has realised many possible applications in areas such as robotics, object recognition, surveillance, manipulation, navigation and interaction. Given the large volumes of data in range images, processing and extracting the relevant information from the images in real time becomes challenging. To achieve this, much research has been conducted in the area of bio-inspired feature extraction which aims to emulate the biological processes used to extract relevant features, reduce redundancy, and process images efficiently. Inspired by the behaviour of biological vision systems, an approach is presented for extracting important features from intensity and range images, using biologically inspired spiking neural networks in order to model aspects of the functional computational capabilities of the visual system.

Linear time series models, such as the autoregressive integrated moving average (ARIMA) model, are among the most popular statistical models used to forecast time series. In recent years non-linear computational models, such as artificial neural networks (ANN), have been shown to outperform traditional linear models when dealing with complex data, like financial time series. This paper proposes a novel hybrid forecasting model which exploits the linear modelling strengths of the ARIMA model, and the flexibility of a self-organising fuzzy neural network (SOFNN). The system's performance is evaluated using several datasets, and our results indicate that a hybrid system is an effective tool for time series forecasting.

Price manipulation refers to the act of using illegal trading behaviour to manually change an equity price with the aim of making profits. With increasing volumes of trading, price manipulation can be extremely damaging to the proper functioning and integrity of capital markets. Effective approaches for analysing and real-time detection of price manipulation are yet to be developed. This paper proposes a novel approach, called Hidden Markov Model with Abnormal States (HMMAS), which models and detects price manipulation activities. Together with the wavelet decomposition for features extraction and Gaussian Mixture Model for Probability Density Function (PDF) construction, the HMMAS model detects price manipulation and identifies the type of the detected manipulation. Evaluation experiments of the model were conducted on six stock tick data from NASDAQ and London Stock Exchange (LSE). The results showed that the proposed HMMAS model can effectively detect price manipulation patterns.

Existing models for document summarization mostly use the similarity between sentences in the document to extract the most salient sentences. The documents as well as the sentences are indexed using traditional term indexing measures, which do not take the context into consideration. Therefore, the sentence similarity values remain independent of the context. In this paper, we propose a context sensitive document indexing model based on the Bernoulli model of randomness. The Bernoulli model of randomness has been used to find the probability of the cooccurrences of two terms in a large corpus. A new approach using the lexical association between terms to give a context sensitive weight to the document terms has been proposed. The resulting indexing weights are used to compute the sentence similarity matrix. The proposed sentence similarity measure has been used with the baseline graph-based ranking models for sentence extraction. Experiments have been conducted over the benchmark DUC data sets and it has been shown that the proposed Bernoulli-based sentence similarity model provides consistent improvements over the baseline IntraLink and UniformLink methods [1].

We propose an evolutionary algorithm to autonomously improve the performances of a robotics skill. The algorithm extends a previously proposed graphical evolutionary skills building approach to allow a robot to autonomously collect use cases where a skill fails and use them to improve the skill. Here we define a computational graph as a generic model to hierarchically represent skills and to modify them. The computational graph makes use of embedded neural networks to create generic skills. We tested our proposed algorithm on a real robot implementing a “move to reach” action. Four experiments show the evolution of the computational graph as it is adapted to solve increasingly complex problems.

Many robots use range data to obtain an almost 3-dimensional description of their environment. Feature driven segmentation of range images has been primarily used for 3D object recognition, and hence the accuracy of the detected features is a prominent issue. Inspired by the structure and behaviour of the human visual system, we present an approach to feature extraction in range data using spiking neural networks and a biologically plausible hexagonal pixel arrangement. Standard digital images are converted into a hexagonal pixel representation and then processed using a spiking neural network with hexagonal shaped receptive fields; this approach is a step towards developing a robotic eye that closely mimics the human eye. The performance is compared with receptive fields implemented on standard rectangular images. Results illustrate that, using hexagonally shaped receptive fields, performance is improved over standard rectangular shaped receptive fields.

This paper presents a position based visual tracking system of a redundant manipulator using a Kinect camera. Kinect camera provides 3-D information of a target object, therefore the control algorithm of the position-based visual servoing (PBVS) can be simplified, as there is no requirement to estimate a 3-D feature point position from the extracted image and the camera model. The Kalman filter is used to predict the target position and velocity. This control method is applied to a calibrated robotic system with eye-to-hand configuration. The stability analysis has been derived and real-time experiments have been carried out using a 7 DOF PowerCube manipulator from Amtec Robotic. The experimental results of both static and moving targets are presented to demonstrate and to verify the proposed position based visual tracking system performance.

In this paper an algorithm is proposed to extract two illuminant invariant chromaticity features from three image sensor responses. The algorithm extracts these chromaticity features at pixel level and therefore can perform well in scenes illuminated with non-uniform illuminant. An approach is proposed to use the algorithm with cameras of unknown sensitivity. The algorithm was tested for separability of perceptually similar colours under the International Commission on Illumination (CIE) standard illuminants and obtained a good performance. It was also tested for colour based object recognition by illuminating objects with typical indoor illuminants and obtained a better performance compared to other existing algorithms investigated in this paper. Finally, the algorithm was tested for skin detection invariant to illuminant, ethnic background and imaging device. In this investigation, daylight scenes under different weather conditions and scenes illuminated by typical indoor illuminants were used. The proposed algorithm gives a better skin detection performance compared to widely used standard colour spaces. Based on the results presented, the proposed illuminant invariant chromaticity space can be used for machine vision applications including illuminant invariant colour based object recognition and skin detection.

A user query for information retrieval (IR) applications may not contain the most appropriate terms (words) as actually intended by the user. This is usually referred to as the term mismatch problem and is a crucial research issue in IR. Using the notion of relevance, we provide a comprehensive theoretical analysis of a parametric query vector, which is assumed to represent the information needs of the user. A lexical association function has been derived analytically using the system relevance criteria. The derivation is further justified using an empirical evidence from the user relevance criteria. Such analytical derivation as presented in this paper provides a proper mathematical framework to the query expansion techniques, which have largely been heuristic in the existing literature. By using the generalized retrieval framework, the proposed query representation model is equally applicable to the vector space model (VSM), Okapi best matching 25 (Okapi BM25), and Language Model (LM). Experiments over various data sets from TREC show that the proposed query representation gives statistically significant improvements over the baseline Okapi BM25 and LM as well as other well-known global query expansion techniques. Empirical results along with the theoretical foundations of the query representation confirm that the proposed model extends the state of the art in global query expansion.

In this paper, a spiking neural network (SNN) architecture to simulate the sound localization ability of the mammalian auditory pathways using the interaural intensity difference cue is presented. The lateral superior olive was the inspiration for the architecture, which required the integration of an auditory periphery (cochlea) model and a model of the medial nucleus of the trapezoid body. The SNN uses leaky integrate-and-fire excitatory and inhibitory spiking neurons, facilitating synapses and receptive fields. Experimentally derived head-related transfer function (HRTF) acoustical data from adult domestic cats were employed to train and validate the localization ability of the architecture, training used the supervised learning algorithm called the remote supervision method to determine the azimuthal angles. The experimental results demonstrate that the architecture performs best when it is localizing high-frequency sound data in agreement with the biology, and also shows a high degree of robustness when the HRTF acoustical data is corrupted by noise.

Brain-computer interface (BCI) technology is a means of communication that allows individuals with severe movement disability to communicate with external assistive devices using the electroencephalogram (EEG) or other brain signals. This paper presents an alternative neural information processing architecture using the Schrödinger wave equation (SWE) for enhancement of the raw EEG signal. The raw EEG signal obtained during the motor imagery (MI) of a BCI user is intrinsically embedded with non-Gaussian noise while the actual signal is still a mystery. The proposed work in the field of recurrent quantum neural network (RQNN) is designed to filter such non-Gaussian noise using an unsupervised learning scheme without making any assumption about the signal type. The proposed learning architecture has been modified to do away with the Hebbian learning associated with the existing RQNN architecture as this learning scheme was found to be unstable for complex signals such as EEG. Besides, this the soliton behaviour of the non-linear SWE was not properly preserved in the existing scheme. The unsupervised learning algorithm proposed in this paper is able to efficiently capture the statistical behaviour of the input signal while making the algorithm robust to parametric sensitivity. This denoised EEG signal is then fed as an input to the feature extractor to obtain the Hjorth features. These features are then used to train a Linear Discriminant Analysis (LDA) classifier. It is shown that the accuracy of the classifier output over the training and the evaluation datasets using the filtered EEG is much higher compared to that using the raw EEG signal. The improvement in classification accuracy computed over nine subjects is found to be statistically significant.

This paper proposes a biologically plausible system for object recognition based on tactile form perception. A spiking neural network, an encoding scheme for converting the input values into spike trains, a method for converting the output spike pattern into reliable features for object recognition and a training approach for the spiking neural network are proposed. Three separate spiking neural networks are used in this recognition system. Three features, based on the output firing pattern of the three networks, are projected onto a three dimensional space. Each class of objects forms a cluster in the three-dimensional feature space. During the training the firing threshold of the hidden layer is modified in such a way that the cluster formed by an object is small and does not overlap with neighbouring clusters. The system has been tested with a number of objects for recognition based on shape. In addition, the system has also been tested for the ability to recognise objects of the same shape but different size. The results show the proposed system gives good performance in recognising objects based on tactile form perception.

This paper outlines the development of a cross-correlation algorithm and a spiking neural network (SNN) for sound localisation based on real sound recorded in a noisy and dynamic environment by a mobile robot. The SNN architecture aims to simulate the sound localisation ability of the mammalian auditory pathways by exploiting the binaural cue of interaural time difference (ITD). The medial superior olive was the inspiration for the SNN architecture which required the integration of an encoding layer which produced biologically realistic spike trains, a model of the bushy cells found in the cochlear nucleus and a supervised learning algorithm. The experimental results demonstrate that biologically inspired sound localisation achieved using a SNN can compare favourably to the more classical technique of cross-correlation.

The raphe nucleus in the brain is the main source of serotonin (5-HT), an important brain chemical in regulating mood, cognition and behavior. This paper presents a spiking neuronal network model of the dorsal region of the raphe nucleus (DRN). We solve the perplexing problem of heterogeneous spiking neuronal behavior observed in the DRN by using an adaptive quadratic integrate-and-fire neuronal model and varying only its membrane potential reset after a spike, suggesting a potential role of certain recovery ionic currents. Specifically, the model can mimic the effects of slow afterhyperpolarization current and control the production of spikes per burst as found in experiments. Our model predicts specific input-output functions of the neurons which can be experimentally tested. Phase-plane analysis confirms their spiking dynamics. By coupling the 5-HT neurons with non-5-HT inhibitory neurons, we show that the DRN neuronal spiking activities recorded in behaving monkeys can generally be reproduced by adopting a feedforward inhibitory network architecture. Our model further predicts a low frequency network oscillation (about 8 Hz) among non-5-HT neurons around the rewarding epoch of a simulated experimental trial, which can be verified through direct recordings in behaving animals. Our computational model of the DRN accounts for the heterogeneous spiking patterns found in experiments, suggests plausible network architecture, and provides model predictions which can be directly tested in experiments. The model conveniently forms the basis for building extended network models to study complex interactions of the 5-HT system with other brain regions.

In a multi-robot system, the coordination and cooperation among the robots determine the effectiveness of task execution. Different centralised and distributed task allocation algorithms have been proposed by researchers. Recently consensus based task allocation has been extensively researched because of its robustness in handling large teams of robots. We propose a new auction and consensus based algorithm for fast task allocation in parallel with task execution. The performance of the proposed algorithm under different conditions is analyzed and compared with other distributed consensus algorithms.

In this paper a biologically inspired spiking neural network based haptic object recognition system is proposed. A number of different encoding schemes to convert the haptic measurements into spike train are proposed and investigated for haptic object recognition and compared with existing encoding schemes. The spiking neural network was trained using a supervised training approach that is based on the steepest descent algorithm. During the training, firing threshold of the hidden layer neurons were modified in such a way that the ability of the system in recognising different objects is maximised. A multiplexing scheme is used to convert the parallel spike train of the hidden layer into a serial stream. To convert the output spike train into a reliable feature that represents the shape of an object, moment of the spikes with respect to a reference time is calculated. A robot hand with three fingers that have the same number of degrees of freedom as the human fingers was used for testing the system. The hand was made to grasp different objects and the joint angles were recorded. These recorded angles were converted into spike train using different encoding schemes and applied as input to the network. Test results show that the performance of the system varies depending on the input encoding scheme and with the best encoring scheme the system can recognise 100% of 7 different objects.

Inspired by the structure and behaviour of the human visual system, we present an approach to edge detection using spiking neural networks and a biologically plausible hexagonal pixel arrangement. Standard digital images are converted into a hexagonal pixel representation and then processed using a spiking neural network with hexagonal shaped receptive fields. The performance is compared with receptive fields implemented on standard rectangular images. Results illustrate that, using hexagonal shaped receptive fields, performance is improved over standard rectangular shaped receptive fields.

In this paper a model selection algorithm for a nonlinear system identification method is proposed to study functional magnetic resonance imaging (fMRI) effective connectivity. Unlike most other methods, this method does not need a pre-defined structure/model for effective connectivity analysis. Instead, it relies on selecting significant nonlinear or linear covariates for the differential equations to describe the mapping relationship between brain output (fMRI response) and input (experiment design). These covariates, as well as their coefficients, are estimated based on a least angle regression (LARS) method. In the implementation of the LARS method, Akaike's information criterion corrected (AICc) algorithm and the leave-one-out (LOO) cross-validation method were employed and compared for model selection. Simulation comparison between the dynamic causal model (DCM), nonlinear identification method, and model selection method for modelling the single-input-single-output (SISO) and multiple-input multiple-output (MIMO) systems were conducted. Results show that the LARS model selection method is faster than DCM and achieves a compact and economic nonlinear model simultaneously. To verify the efficacy of the proposed approach, an analysis of the dorsal and ventral visual pathway networks was carried out based on three real datasets. The results show that LARS can be used for model selection in an fMRI effective connectivity study with phase-encoded, standard block, and random block designs. It is also shown that the LOO cross-validation method for nonlinear model selection has less residual sum squares than the AICc algorithm for the study.

This paper presents a distributed fuzzy proportional control system for a vision guided redundant robot manipulator. Firstly, the Takagi Sugeno (TS) fuzzy algorithm is used to model analytical Image Based Visual Servoing (IBVS) using shape moments by offline learning. This control method is applied to an uncalibrated robotic system with eye-in-hand visual feedback. The system is able to track a moving object through a variety of motions and maintain the object's image features in a desired position in the image plane without a priori knowledge of the robot kinematic, camera calibration and inverse Jacobian. The experimental results of both static and moving targets using the 7 DOF PowerCube manipulator from Amtec Robotic show and verify its performance in a realtime application.

This study investigates the computing capabilities and potential applications of neural oscillators to grey scale and colour image segmentation, an important task in image understanding and object recognition. A proposed neural system that combines the synergy between neural oscillators and Kohonen self-organising maps (SOM) is presented. Colour image segmentation is achieved through temporal synchronisation of neural oscillators that are mapped to pixels of the same object. Neurons are organised in a two-dimensional grid and are locally connected through excitatory connections and globally connected to a common inhibitor. Self-organising maps form the basis of a colour reduction system whose output is fed to a 2D grid of neural oscillators such as each neuron is mapped to a pixel of the input image. Both chromatic and local spatial features are used. The proposed system is simulated in Matlab and its demonstration on real world colour images shows promising results and the emergence of a new bio-inspired approach for colour image segmentation.

In this work the self-organizing fuzzy neural network (SOFNN) is employed to create an accurate and easily calibrated approach to multiple-step-ahead prediction for the NN5 forecasting competition 2008. The competition dataset consists of 111 daily empirical time series of cash-machine withdrawals. The objective for the competition was to forecast future transactions up to 56 days ahead with the highest prediction accuracy using a single methodology. The SOFNN is a highly efficient and accurate algorithm for time series-prediction which learns from data incrementally and can autonomously adapt its structure in the learning process to cope with drifts in the data dynamics. It can also modify its architecture autonomously to suit different prediction horizons, embedding dimensions and time lags. Standard neural networks(NNs) and autoregressive(AR) models are employed as benchmarks for comparison. It is shown through a statistical analysis of the results, that the SOFNN significantly outperforms the NN and AR methods.

Recent work has shown that combining prediction based preprocessing based on neural-time-series-prediction-preprocessing (NTSPP) along with spectral filtering (SF) and common-spatial patterns (CSP) can significantly improve the performance of a motor imagery based brain-computer interface (BCI) involving two classes. This paper illustrates how these performance improvements can be extended to a 4 class motor imagery BCI with between 2 and 22 channels. The results show that this combination of preprocessing techniques can significantly outperform any of methods operating independently and that NTSPP can reduce the number of electrodes required based on a comparison of results from 2, 3 and multichannel data.

This paper presents an approach that permits the effective hardware realization of a novel Evolvable Spiking Neural Network (ESNN) paradigm on Field Programmable Gate Arrays (FPGAs). The ESNN possesses a hybrid learning algorithm that consists of a Spike Timing Dependent Plasticity (STDP) mechanism fused with a Genetic Algorithm (GA). The design and implementation direction utilizes the latest advancements in FPGA technology to provide a partitioned hardware/software co-design solution. The approach achieves the maximum FPGA flexibility obtainable for the ESNN paradigm. The algorithm was applied as an embedded intelligent system robotic controller to solve an autonomous navigation and obstacle avoidance problem.

This paper proposes a supervised training algorithm for Spiking Neural Networks (SNNs) which modifies the Spike Timing Dependent Plasticity (STDP)learning rule to support both local and network level training with multiple synaptic connections and axonal delays. The training algorithm applies the rule to two and three layer SNNs, and is benchmarked using the Iris and Wisconsin Breast Cancer (WBC) data sets. The effectiveness of hidden layer dynamic threshold neurons is also investigated and results are presented.

Spiking neural networks (SNNs) are an emerging computing paradigm that attempt to model the biological functions of the human brain. However, as networks approach the biological scale with significantly large numbers of neurons, software simulations face the problem of scalability and increasing computation times. Thus, numerous researchers have targeted hardware implementations in an attempt to more closely replicate the parallel processing capabilities of biological networks. Reconfigurable hardware is seen as a particularly viable platform for attempting to replicate to some degree the natural plasticity and flexibility of the human brain. This paper presents a scalable FPGA based implementation approach that facilitates the accelerated emulation of large-scale SNNs. The approach is validated using a SNN-based edge detection application where an order of magnitude speed performance increase was observed in comparison to a software equivalent implementation.

This paper introduces a number of modifications to the learning algorithm of the self-organizing fuzzy neural network (SOFNN) to improve computational efficiency. It is shown that the modified SOFNN favorably compares to other evolving fuzzy systems in terms of accuracy and structural complexity. An analysis of the SOFNN's effectiveness when applied in an electroencephalogram (EEG)-based brain-computer interface (BCI) involving the neural-time-series-prediction-preprocessing (NTSPP) framework is also presented, where a sensitivity analysis (SA) of the SOFNN hyperparameters was performed using EEG data recorded from three subjects during left/right-motor-imagery-based BCI experiments. The aim of this one-time SA was to eliminate the need to choose subject- and signal-specific hyperparameters for the SOFNN and thus apply the SOFNN in the NTSPP framework as a parameterless self-organizing framework for EEG preprocessing. The results indicate that a general set of NTSPP parameters chosen via the SA provide the best results when tested in a BCI system. Therefore, with this general set of SOFNN parameters and its self-organizing structure, in conjunction with parameterless feature extraction and linear discriminant classification, a fully parameterless BCI that lends itself well to autonomous adaptation is realizable.

Recent research has shown that neural networks (NNs) or self-organizing fuzzy NNs (SOFNNs) can enhance the separability of motor imagery altered electroencephalogram (EEG) for brain-computer interface (BCI) systems. This is achieved via the neural-time-series-prediction-preprocessing (NTSPP) framework where SOFNN prediction models are trained to specialize in predicting the EEG time-series recorded from different EEG channels whilst subjects perform various mental tasks. Features are extracted from the predicted signals produced by the SOFNN and it has been shown that these features are easier to classify than those extracted from the original EEG. Previous work was based on a two class BCI. This paper presents an analysis of the NTSPP framework when extended to operate in a multiclass BCI system. In mutliclass systems normally multiple EEG channels are used and a significant amount of subject-specific parameters and EEG channels are investigated. This paper demonstrates how the SOFNN-based NTSPP, tested in conjunction with three different feature extraction procedures and different linear discriminant and support vector machine (SVM) classifiers, is effective in improving the performance of a multiclass BCI system, even with a low number of standardly positioned electrodes and no subject-specific parameter tuning.

Sound localisation is defined as the ability to identify the position of a sound source. The brain employs two cues to achieve this functionality for the horizontal plane, interaural time difference (ITD) by means of neurons in the medial superior olive (MSO) and interaural intensity difference (IID) by neurons of the lateral superior olive (LSO), both located in the superior olivary complex of the auditory pathway. This paper presents spiking neuron architectures of the MSO and LSO. An implementation of the Jeffress model using spiking neurons is presented as a representation of the MSO, while a spiking neuron architecture showing how neurons of the medial nucleus of the trapezoid body interact with LSO neurons to determine the azimuthal angle is discussed. Experimental results to support this work are presented.

Type-2 (T2) fuzzy logic (FL) systems (T2FLSs) have shown a remarkable potential in dealing with uncertain data resulting from real-world systems with non-stationary characteristics. This paper reports on novel developments in interval T2FLS (IT2FLS) classifier design methodology so that system non-stationarities can be effectively handled. In general, the approach presented here rests on a general concept of two-stage FLS design in which an initial rule base structure is first initialized and then system parameters are globally optimized. The proposed incremental enhancements of existing fuzzy techniques, adopted from the area of conventional type-1 (T1) FL, are heuristic in nature. The IT2FLS design methods have been empirically verified in this work in the realm of pattern recognition. In particular, the potential and the suitability of IT2FLS to the problem of classification of motor imagery (MI) related patterns in electroencephalogram (EEG) recordings has been investigated. The outcome of this study bears direct relevance to the development of EEG-based brain-computer interfaces (BCIs) since the problem under examination poses a major difficulty for the state-of-the-art BCI methods. The IT2FLS classifier is evaluated in this work on multi-session EEG data sets in the framework of an off-line BCI. Its performance is quantified in terms of the classification accuracy (CA) rates and has been found to be favorable to that of analogous systems employing a conventional T1FLS, along with linear discriminant analysis (LDA) and support vector machine (SVM), commonly utilized in MI-based BCI systems.

The significance of the initialization procedure in the development of Type-2 fuzzy logic (T2FL) system-based classifiers should be highlighted considering their intrinsically non-linear nature. Initial structure identification has been recognized as a crucial stage in the design of an interval T2FL (IT2FL) classifier utilized in the framework of electroencephalogram (EEG)-based brain -computer interface (BCI). In conjunction with an efficient gradient-based learning algorithm it has allowed for robust exploitation of T2FL's capabilities to effectively handle uncertainties inherently associated with changing dynamics of electrical brain activity. This paper builds on the previous experiences in tackling the problem of inter-session classification of motor imagery (MI)-related EEG patterns. The major contribution of this work is an empirical investigation of the concept of support vector (SV) learning applied to structure identification of the IT2FL classifier. The SV-enhanced initialization scheme is found to compare favorably to both an arbitrary initialization and the clustering approach utilized in the preceding work in terms of the inter-session BCI classification performance of the fully trained IT2FLS evaluated on three subjects.

Autonomic computing is gradually becoming accepted as a viable approach to achieving self- management in systems and networks, with the goal of lessening the impact of the complexity crisis on the computing industry. The authors propose the integration of high level self-organisation features into the Application Directed Adaptive Framework (ADAF) which, when embedded into software applications, enables those applications to exhibit autonomic behaviour. This paper briefly discusses the infrastructure of ADAF and demonstrates two self-managing capabilities, namely self- monitoring and self-diagnosis.

When symbolic AI approaches are applied to handle continuous valued attributes, there is a requirement to transform the continuous attribute values to symbolic data. In this paper, a novel distribution-index-based discretizer is proposed for such a transformation. Based on definitions of dichotomic entropy and a compound distributional index, a simple criterion is applied to discretize continuous attributes adaptively. The dichotomic entropy indicates the homogeneity degree of the decision value distribution, and is applied to determine the best splitting point. The compound distributional index combines both the homogeneity degrees of attribute value distributions and the decision value distribution, and is applied to determine which interval should be split further; thus, a potentially improved solution of the discretization problem can be found efficiently. Based on multiple reducts in rough set theory, a multiknowledge approach can attain high decision accuracy for information systems with a large number of attributes and missing values. In this paper, our discretizer is combined with the multiknowledge approach to further improve decision accuracy for information systems with continuous attributes. Experimental results on benchmark data sets show that the new discretizer can improve not only the multiknowledge approach, but also the naive Bayes classifier and the C5.0 tree

In this paper an area efficient multiplier-less hardware architecture is proposed for the implementation of an integrate- and-fire SNN model. The proposed architecture is intended for large scale implementation on a single FPGA. A modular design is proposed in order to make it flexible. Synaptic multiplication is performed with a simple AND gate, and pulses from different synapses are added together at different times, replicating the accumulation of synaptic inputs for the membrane potential. In order to introduce non-linearity into the membrane potential a normalized random number is introduced to this state variable. The proposed architecture uses spike trains as an input much like those in real networks

This paper presents a number of enhancements to the self-organizing fuzzy neural network (SOFNN). Firstly, the SOFNN is described and a modification to the learning algorithm to improve computational efficiency is introduced. Secondly, a sensitivity analysis (SA) of the predefined SOFNN parameters is presented using electroencephalogram (EEG) data recorded from three subjects during left/right motor imagery-based brain-computer interface (BCI) experiments. This SA was carried out to determine if a general set of parameters could be used for predicting various non-stationary EEG time-series dynamics for multiple subjects. The SOFNN modifications significantly enhance computational efficiency and the SA results suggest that it may be possible to select a general set of parameters for different motor imagery-based EEG signals thus potentially enhancing the SOFNNs autonomy for application in a BCI.

This paper presents an extension of previous work whereby the Spike Timing Dependant Plasticity (STDP) rule was used to train a two layer Spiking Neural Network (SNN). In that work a supervised training algorithm was developed using an STDP based rule that affected weights both locally and at network level. This work extends the rule to a three layer network with multiple inter-neuron excitatory synaptic connections and associated delays. The network utilises dynamic thresholds to facilitate an association between spatial patterns in the input data and classes. The algorithm is benchmarked using nonlinearly separable classification problems and results show that the three layer network exhibits a significant improvement over the two layer.

A novel inter-neuron communications method for increased scalability of Spiking Neural Networks (SNNs) is presented. Capacitive coupling is used as the communication medium with spike functions replaced by oscillatory bursts. The dependency of the coupling signals between neuron layers as a function of neuron density, frequency and track loading is predicted. High Q decoding filters and accurate frequency matching between transmitting neurons and receiving synapses was achieved using band pass filters. Micro-Electro-Mechanical Systems (MEMS) were used in the filter circuits and burst oscillators because of their high Q values. The method of fabricating the on-chip coupling capacitors and associated oscillator/filter circuits is discussed.

A novel hybrid learning algorithm based on a genetic algorithm to design a growing fuzzy neural network, named self-organizing fuzzy neural network based on genetic algorithms (SOFNNGA), to implement Takagi-Sugeno (TS) type fuzzy models is proposed in this paper. A new adding method based on geometric growing criterion and the epsiv-completeness of fuzzy rules is first used to generate the initial structure. Then a hybrid algorithm based on genetic algorithms, backpropagation, and recursive least squares estimation is used to adjust all parameters including the number of fuzzy rules. This has two steps: First, the linear parameter matrix is adjusted, and second, the centers and widths of all membership functions are modified. The GA is introduced to identify the least important neurons, i.e., the least important fuzzy rules. Simulations are presented to illustrate the performance of the proposed algorithm

A self-organizing computing network based on concepts of fuzzy conditions, beliefs, probabilities, and neural networks is proposed for decision-making in intelligent systems which are required to handle data sets with a diversity of data types. A sense-function with a sense-range and fuzzy edges is defined as a transfer function for connections from the input layer to the hidden layer in the network. By generating hidden cells and adjusting the parameters of the sense-functions, the network self-organizes and adapts to a training set. Computing cells in the input layer are designed as data converters so that the network can deal with both symbolic data and numeric data. Hidden computing cells in the network can be explained via fuzzy rules in a similar manner to those in fuzzy neural networks. The values in the output layer can be explained as a belief distribution over a decision space. The final decision is made by means of the winner-take-all rule. The approach was applied to a series of the benchmark data sets with a diversity of data types and comparative results obtained. Based on these results, the suitability of a range of data types for processing by different intelligent techniques was analyzed, and the results show that the proposed approach is better than other approaches for decision-making in information systems with mixed data types.

In nano-architectures, transistor counts place extreme demands on interconnection resources. The chronic problem of interconnect area versus device area becomes more acute than it currently is. Even with multilayering of conductors there may still be attenuation and propagation delay issues which at the extremes of nano-architecture severely limit performance. When reconfigurable devices such as field programmable gate arrays (FPGAs) are considered the issue of interconnection resources becomes even more acute due to the inherent redundancy already apparent in existing devices. The authors propose the use of a multiple valued signal system to increase functional density whilst reducing the interconnection resource overhead in FPGA based nano-architectures.

An algorithm inspired by the biological phenomenon of symbiosis is presented in this paper. The genetic diversity obtained from the creation of symbiotic relationships is investigated and the symbiosis algorithm is applied to a robotic forward kinematics control problem. Compared with other evolutionary optimisation techniques, the symbiosis algorithm is shown to be an effective paradigm in discovering optimal solutions. Genetic diversity was found to be maintained in the absence of conventional genetic operators.

In this paper, an evolving learning mechanism is proposed for general computing network model to make decisions in intelligent systems. The novel mechanism is performed by means of a set of computing cell operations such as self-generation, growth, self-division, and death. Under the mechanism, a computing network grows up to a mature network. A hidden cell in the network is defined as a condition matching-unit in response to a fuzzy sub-superspace in multiple-dimension input superspace. A sense-function is defined to represent connections from a hidden cell to input cells. The range and edge vagueness of the sense-function are determined by evolving learning mechanism when sample instances are presented to the network. This network is able to learn from a very few training instances to make decisions for unseen instances. The benchmark data sets from the UCI machine learning repository are applied to test the network and comparable results are obtained.

Threshold logic gates (TLGs) based on resonant tunneling diodes (RTDs) allow the implementation of simple McCulloch-Pitts (MCP) neuron models. Whilst they have been demonstrated implementing a wide range of binary logic circuits the gates are also theoretically capable of implementing artificial neural networks. The conventional operation of RTD based logic circuits with many cascaded stages has inherent disadvantages associated with the requirement of an evaluation period at each stage and the resultant use of multiple clocks. The authors propose that highly parallel structures similar to those found in artificial neural networks are better suited to RTD based TLGs and offer the way forward for future large scale designs.

Increasingly web based distance education engineering courses are on offer, augmented by the provision of remote experimentation laboratories facilitating distant access to campus based physical resources. The design and implementation of effective and usable remote experimentation facilities poses unique challenges given the inherent complexities of the learning environment and the constraints imposed by the delivery medium. Developments in recent years have addressed many of these issues. However autonomous learning environments by their very nature offer minimal educator assistance and from a students perspective it is inevitable that at some stage of the experimental process, context specific help will be required. This paper seeks to address this issue in the context of remote experimentation for embedded systems and presents a distributed communications application implemented using a web services/.NET Remoting framework for an adaptive intelligent learning environment.

Constant innovation and product evolution in the area of embedded systems necessitates educational institutions and other training providers to continuingly reassess the content and delivery of engineering curricula. Increasingly web based distance education courses are on offer, augmented by the provision of remote experimentation laboratories facilitating distant access to campus based physical resources. The design and implementation of effective and usable remote experimentation facilities poses unique challenges given the inherent complexities of the learning environment and the constraints imposed by the delivery medium. Developments in recent years have addressed many of these issues. However autonomous learning environments by their very nature offer minimal educator assistance. This paper seeks to address this issue in the context of remote experimentation for embedded systems and demonstrates a distributed communications application implemented using a web services/.NET Remoting framework for an adaptive intelligent learning environment for remote experimentation.

Abstract. This paper details a new object-oriented methodology that permits a unified modelling language (UML) behavioural representation of analogue circuits at system level. The proposed method demonstrates a novel approach to the problem of behavioural representation of an analogue topology, by constructing a consistent set of rules for automated mapping of the UML model to a VHDL-AMS specification. The VHDL-AMS specification enables behavioural simulation of the UML model and the methodology is validated using an analogue subsystem level application.

The hardware--software (HW--SW) partitioning of applications to dynamically reconfigurable embedded systems allows for customization of their hardware resources during run-time to meet the demands of executing applications. The run-time reconfiguration (RTR) of such systems can have an impact on the HW--SW partitioning strategy and the system performance. It is therefore important to consider approaches to optimally reduce the RTR overhead during the HW--SW partitioning stage. In order to examine potential benefits in performance, it is necessary to develop a method to model and evaluate the RTR. In this paper, a novel method of modeling and evaluating such RTR-reduced HW--SW partitions is presented. The techniques of computation-reconfiguration overlap and the retention of circuitry between reconfigurations are used within this model to explore the possibilities of RTR reduction. The integration of this model into the authors' current genetic-algorithm-driven HW--SW partitioner is also presented, with two applications used to illustrate the benefits of RTR-reduced exploration during HW--SW partitioning.

Embedded systems are an integral part of industrial and domestic electronics necessitating educational institutions and other training providers to offer advanced embedded systems courses. Effective teaching in this area requires an interdependent approach combining theoretical material underpinned by practical laboratory experiments and individualized tutoring. Increasingly the courses on offer include web based distance education packages augmented by the provision of remote experimentation laboratories facilitating distant access to campus based physical resources. The degree of functionality and level of user access to remote laboratories has evolved in recent years expedited by advances in web applications and technologies. This paper details our recent work in this area focusing on the development of architecture for an integrated remote experimentation laboratory designed to meet the unique requirements of a complex, dynamic learning environment.

A novel hybrid algorithm based on the genetic algorithm, named self-organizing fuzzy neural network based on genetic algorithm (SOFNNGA), is proposed to design a fuzzy neural network to implement Takagi-Sugeno (TS) type fuzzy models in this paper. A new adding method based on geometric growing criterion and the /spl epsiv/-completeness of fuzzy rules is used to generate the initial structure firstly. Then a hybrid algorithm based on genetic algorithms, backpropagation, and recursive least squares estimation is used to adjust all parameters, which has two steps: first, adjusting the parameter matrix, and second, centers and widths of all membership functions are modified. A simulation for a benchmark problem is presented to illustrate the performance of the proposed algorithm.

The use of laboratory experiments is a critically important aspect of engineering education where experience has shown that a complementary approach combining theoretical and practical exercises is vital for effective learning. Increasingly, teaching institutions are offering Web based remote access to distant laboratories as part of an overall e-learning strategy. The design and implementation of effective and usable remote experimentation facilities poses unique challenges given the inherent complexities of the learning environment and the constraints imposed by the delivery medium. Developments in recent years have addressed many of these issues. However autonomous learning environments by their very nature offer minimal educator assistance and from a students perspective it is inevitable that at some stage of the experimental process, context specific help will be required. We address this issue in the context of remote experimentation for embedded systems and present an adaptive intelligent learning environment with intelligent user help.

The use of laboratory experiments is a critically important aspect of engineering education. Experience in teaching has shown that a complementary approach combining theoretical and practical exercises is vital for effective learning. Increasingly, teaching institutions are offering remote access to distant laboratories as part of an overall e-learning strategy. Remote experimentation provided as part of a Web-based learning approach affords a number of critical benefits, allowing flexible access to on-campus resources free of time or geographical constraints. In the context of distance learning it can be the only realistic method of performing experiments. However adapting and redeveloping existing software and hardware resources to this purpose is both time consuming and expensive. An innovative simple approach to this problem is presented in this paper. The remote desktop method offered here allows students to have full access to existing on-campus laboratory resources remotely but without substantial implementation and development overheads incurred by the providers. This is achieved through increased functionality and extended utilisation of existing assets.

In an increasingly competitive marketplace system complexity continues to grow, but time-to-market and lifecycle are reducing. The purpose of fault diagnosis is the isolation of faults on defective systems, a task requiring a high skill set. This has driven the need for automated diagnostic tools. Over the last two decades, automated diagnosis has been an active research area, but the industrial acceptance of these techniques, particularly in cost-sensitive areas, has not been high. This paper reviews this research, primarily covering rule-based, model-based, and case-based approaches and applications. Future research directions are finally examined, with a concentration on issues, which may lead to a greater acceptance of automated diagnosis.

The authors propose an evolutionary algorithm (EA) approach to hardware-software partitioning and performance estimation of dynamically reconfigurable embedded systems. A demonstrative application is used to show the effectiveness of the use of GAs to achieve hardware-software partitions with maximum speedup.