DARPA neural network study final report /

Mode of access: Internet.

Tracking the states of a nonlinear and nonstationary system in the weight-space of artificial neural networks

We propose a novel interpretation and usage of Neural Network (NN) in modeling physiological signals, which are allowed to be nonlinear and/or nonstationary. The method consists of training a NN for the k-step prediction of a physiological signal, and then examining the connection-weight-space (CWS) of the NN to extract information about the signal generator mechanism. We de. ne a novel feature, Normalized Vector Separation (gamma(ij)), to measure the separation of two arbitrary states i and j in the CWS and use it to track the state changes of the generating system. The performance of the method is examined via synthetic signals and clinical EEG. Synthetic data indicates that gamma(ij) can track the system down to a SNR of 3.5 dB. Clinical data obtained from three patients undergoing carotid endarterectomy of the brain showed that EEG could be modeled (within a root-means-squared-error of 0.01) by the proposed method, and the blood perfusion state of the brain could be monitored via gamma(ij), with small NNs having no more than 21 connection weight altogether.

A novel neural network technique for exploratory data analysis

This paper surveys the context of feature extraction by neural network approaches, and compares and contrasts their behaviour as prospective data visualisation tools in a real world problem. We also introduce and discuss a hybrid approach which allows us to control the degree of discriminatory and topographic information in the extracted feature space.

A novel silicon membrane-based biosensing platform using distributive sensing strategy and artificial neural networks for feature analysis

A novel biosensing system based on a micromachined rectangular silicon membrane is proposed and investigated in this paper. A distributive sensing scheme is designed to monitor the dynamics of the sensing structure. An artificial neural network is used to process the measured data and to identify cell presence and density. Without specifying any particular bio-application, the investigation is mainly concentrated on the performance testing of this kind of biosensor as a general biosensing platform. The biosensing experiments on the microfabricated membranes involve seeding different cell densities onto the sensing surface of membrane, and measuring the corresponding dynamics information of each tested silicon membrane in the form of a series of frequency response functions (FRFs). All of those experiments are carried out in cell culture medium to simulate a practical working environment. The EA.hy 926 endothelial cell lines are chosen in this paper for the bio-experiments. The EA.hy 926 endothelial cell lines represent a particular class of biological particles that have irregular shapes, non-uniform density and uncertain growth behaviour, which are difficult to monitor using the traditional biosensors. The final predicted results reveal that the methodology of a neural-network based algorithm to perform the feature identification of cells from distributive sensory measurement has great potential in biosensing applications.

The impact of architecture on the performance of artificial neural networks

A number of researchers have investigated the impact of network architecture on the performance of artificial neural networks. Particular attention has been paid to the impact on the performance of the multi-layer perceptron of architectural issues, and the use of various strategies to attain an optimal network structure. However, there are still perceived limitations with the multi-layer perceptron and networks that employ a different architecture to the multi-layer perceptron have gained in popularity in recent years, particularly, networks that implement a more localised solution, where the solution in one area of the problem space does not impact, or has a minimal impact, on other areas of the space. In this study, we discuss the major architectural issues affecting the performance of a multi-layer perceptron, before moving on to examine in detail the performance of a new localised network, namely the bumptree. The work presented here examines the impact on the performance of artificial neural networks of employing alternative networks to the long established multi-layer perceptron. In particular, networks that impose a solution where the impact of each parameter in the final network architecture has a localised impact on the problem space being modelled are examined. The alternatives examined are the radial basis function and bumptree neural networks, and the impact of architectural issues on the performance of these networks is examined. Particular attention is paid to the bumptree, with new techniques for both developing the bumptree structure and employing this structure to classify patterns being examined.

The application of artificial neural networks to the interpretation and classification of freshwater benthic invertebrate communities

This thesis presents a thorough and principled investigation into the application of artificial neural networks to the biological monitoring of freshwater. It contains original ideas on the classification and interpretation of benthic macroinvertebrates, and aims to demonstrate their superiority over the biotic systems currently used in the UK to report river water quality. The conceptual basis of a new biological classification system is described, and a full review and analysis of a number of river data sets is presented. The biological classification is compared to the common biotic systems using data from the Upper Trent catchment. This data contained 292 expertly classified invertebrate samples identified to mixed taxonomic levels. The neural network experimental work concentrates on the classification of the invertebrate samples into biological class, where only a subset of the sample is used to form the classification. Other experimentation is conducted into the identification of novel input samples, the classification of samples from different biotopes and the use of prior information in the neural network models. The biological classification is shown to provide an intuitive interpretation of a graphical representation, generated without reference to the class labels, of the Upper Trent data. The selection of key indicator taxa is considered using three different approaches; one novel, one from information theory and one from classical statistical methods. Good indicators of quality class based on these analyses are found to be in good agreement with those chosen by a domain expert. The change in information associated with different levels of identification and enumeration of taxa is quantified. The feasibility of using neural network classifiers and predictors to develop numeric criteria for the biological assessment of sediment contamination in the Great Lakes is also investigated.

Training a Linear Neural Network with a Stable LSP Solution for Jamming Cancellation

Two jamming cancellation algorithms are developed based on a stable solution of least squares problem (LSP) provided by regularization. They are based on filtered singular value decomposition (SVD) and modifications of the Greville formula. Both algorithms allow an efficient hardware implementation. Testing results on artificial data modeling difficult real-world situations are also provided.

Self-Learning Fuzzy Spiking Neural Network as a Nonlinear Pulse-Position Threshold Detection Dynamic System Based on Second-Order Critically Damped Response Units

Architecture and learning algorithm of self-learning spiking neural network in fuzzy clustering task are outlined. Fuzzy receptive neurons for pulse-position transformation of input data are considered. It is proposed to treat a spiking neural network in terms of classical automatic control theory apparatus based on the Laplace transform. It is shown that synapse functioning can be easily modeled by a second order damped response unit. Spiking neuron soma is presented as a threshold detection unit. Thus, the proposed fuzzy spiking neural network is an analog-digital nonlinear pulse-position dynamic system. It is demonstrated how fuzzy probabilistic and possibilistic clustering approaches can be implemented on the base of the presented spiking neural network.

Artificial neural networks for surface roughness prediction when face milling Al 7075-T7351

In this work, different artificial neural networks (ANN) are developed for the prediction of surface roughness (R a) values in Al alloy 7075-T7351 after face milling machining process. The radial base (RBNN), feed forward (FFNN), and generalized regression (GRNN) networks were selected, and the data used for training these networks were derived from experiments conducted using a high-speed milling machine. The Taguchi design of experiment was applied to reduce the time and cost of the experiments. From this study, the performance of each ANN used in this research was measured with the mean square error percentage and it was observed that FFNN achieved the best results. Also the Pearson correlation coefficient was calculated to analyze the correlation between the five inputs (cutting speed, feed per tooth, axial depth of cut, chip°s width, and chip°s thickness) selected for the network with the selected output (surface roughness). Results showed a strong correlation between the chip thickness and the surface roughness followed by the cutting speed. © ASM International.

A new unsupervised convolutional neural network model for Chinese scene text detection

As one of the most popular deep learning models, convolution neural network (CNN) has achieved huge success in image information extraction. Traditionally CNN is trained by supervised learning method with labeled data and used as a classifier by adding a classification layer in the end. Its capability of extracting image features is largely limited due to the difficulty of setting up a large training dataset. In this paper, we propose a new unsupervised learning CNN model, which uses a so-called convolutional sparse auto-encoder (CSAE) algorithm pre-Train the CNN. Instead of using labeled natural images for CNN training, the CSAE algorithm can be used to train the CNN with unlabeled artificial images, which enables easy expansion of training data and unsupervised learning. The CSAE algorithm is especially designed for extracting complex features from specific objects such as Chinese characters. After the features of articficial images are extracted by the CSAE algorithm, the learned parameters are used to initialize the first CNN convolutional layer, and then the CNN model is fine-Trained by scene image patches with a linear classifier. The new CNN model is applied to Chinese scene text detection and is evaluated with a multilingual image dataset, which labels Chinese, English and numerals texts separately. More than 10% detection precision gain is observed over two CNN models.

A novel neural network based system for assessing risks associated with information technology security breaches

Security remains a top priority for organizations as their information systems continue to be plagued by security breaches. This dissertation developed a unique approach to assess the security risks associated with information systems based on dynamic neural network architecture. The risks that are considered encompass the production computing environment and the client machine environment. The risks are established as metrics that define how susceptible each of the computing environments is to security breaches. ^ The merit of the approach developed in this dissertation is based on the design and implementation of Artificial Neural Networks to assess the risks in the computing and client machine environments. The datasets that were utilized in the implementation and validation of the model were obtained from business organizations using a web survey tool hosted by Microsoft. This site was designed as a host site for anonymous surveys that were devised specifically as part of this dissertation. Microsoft customers can login to the website and submit their responses to the questionnaire. ^ This work asserted that security in information systems is not dependent exclusively on technology but rather on the triumvirate people, process and technology. The questionnaire and consequently the developed neural network architecture accounted for all three key factors that impact information systems security. ^ As part of the study, a methodology on how to develop, train and validate such a predictive model was devised and successfully deployed. This methodology prescribed how to determine the optimal topology, activation function, and associated parameters for this security based scenario. The assessment of the effects of security breaches to the information systems has traditionally been post-mortem whereas this dissertation provided a predictive solution where organizations can determine how susceptible their environments are to security breaches in a proactive way. ^

An investigation of location effect on trip production using artificial neural networks

As traffic congestion exuberates and new roadway construction is severely constrained because of limited availability of land, high cost of land acquisition, and communities' opposition to the building of major roads, new solutions have to be sought to either make roadway use more efficient or reduce travel demand. There is a general agreement that travel demand is affected by land use patterns. However, traditional aggregate four-step models, which are the prevailing modeling approach presently, assume that traffic condition will not affect people's decision on whether to make a trip or not when trip generation is estimated. Existing survey data indicate, however, that differences exist in trip rates for different geographic areas. The reasons for such differences have not been carefully studied, and the success of quantifying the influence of land use on travel demand beyond employment, households, and their characteristics has been limited to be useful to the traditional four-step models. There may be a number of reasons, such as that the representation of influence of land use on travel demand is aggregated and is not explicit and that land use variables such as density and mix and accessibility as measured by travel time and congestion have not been adequately considered. This research employs the artificial neural network technique to investigate the potential effects of land use and accessibility on trip productions. Sixty two variables that may potentially influence trip production are studied. These variables include demographic, socioeconomic, land use and accessibility variables. Different architectures of ANN models are tested. Sensitivity analysis of the models shows that land use does have an effect on trip production, so does traffic condition. The ANN models are compared with linear regression models and cross-classification models using the same data. The results show that ANN models are better than the linear regression models and cross-classification models in terms of RMSE. Future work may focus on finding a representation of traffic condition with existing network data and population data which might be available when the variables are needed to in prediction.

A research platform for artificial neural networks with applications in pediatric epilepsy

This dissertation established a state-of-the-art programming tool for designing and training artificial neural networks (ANNs) and showed its applicability to brain research. The developed tool, called NeuralStudio, allows users without programming skills to conduct studies based on ANNs in a powerful and very user friendly interface. A series of unique features has been implemented in NeuralStudio, such as ROC analysis, cross-validation, network averaging, topology optimization, and optimization of the activation function’s slopes. It also included a Support Vector Machines module for comparison purposes. Once the tool was fully developed, it was applied to two studies in brain research. In the first study, the goal was to create and train an ANN to detect epileptic seizures from subdural EEG. This analysis involved extracting features from the spectral power in the gamma frequencies. In the second application, a unique method was devised to link EEG recordings to epileptic and nonepileptic subjects. The contribution of this method consisted of developing a descriptor matrix that can be used to represent any EEG file regarding its duration and the number of electrodes. The first study showed that the inter-electrode mean of the spectral power in the gamma frequencies and its duration above a specific threshold performs better than the other frequencies in seizure detection, exhibiting an accuracy of 95.90%, a sensitivity of 92.59%, and a specificity of 96.84%. The second study yielded that Hjorth’s parameter activity is sufficient to accurately relate EEG to epileptic and non-epileptic subjects. After testing, accuracy, sensitivity and specificity of the classifier were all above 0.9667. Statistical tests measured the superiority of activity at over 99.99 % certainty. It was demonstrated that (1) the spectral power in the gamma frequencies is highly effective in locating seizures from EEG and (2) activity can be used to link EEG recordings to epileptic and non-epileptic subjects. These two studies required high computational load and could be addressed thanks to NeuralStudio. From a medical perspective, both methods proved the merits of NeuralStudio in brain research applications. For its outstanding features, NeuralStudio has been recently awarded a patent (US patent No. 7502763).

A Research Platform for Artificial Neural Networks with Applications in Pediatric Epilepsy

This dissertation established a state-of-the-art programming tool for designing and training artificial neural networks (ANNs) and showed its applicability to brain research. The developed tool, called NeuralStudio, allows users without programming skills to conduct studies based on ANNs in a powerful and very user friendly interface. A series of unique features has been implemented in NeuralStudio, such as ROC analysis, cross-validation, network averaging, topology optimization, and optimization of the activation function’s slopes. It also included a Support Vector Machines module for comparison purposes. Once the tool was fully developed, it was applied to two studies in brain research. In the first study, the goal was to create and train an ANN to detect epileptic seizures from subdural EEG. This analysis involved extracting features from the spectral power in the gamma frequencies. In the second application, a unique method was devised to link EEG recordings to epileptic and non-epileptic subjects. The contribution of this method consisted of developing a descriptor matrix that can be used to represent any EEG file regarding its duration and the number of electrodes. The first study showed that the inter-electrode mean of the spectral power in the gamma frequencies and its duration above a specific threshold performs better than the other frequencies in seizure detection, exhibiting an accuracy of 95.90%, a sensitivity of 92.59%, and a specificity of 96.84%. The second study yielded that Hjorth’s parameter activity is sufficient to accurately relate EEG to epileptic and non-epileptic subjects. After testing, accuracy, sensitivity and specificity of the classifier were all above 0.9667. Statistical tests measured the superiority of activity at over 99.99 % certainty. It was demonstrated that 1) the spectral power in the gamma frequencies is highly effective in locating seizures from EEG and 2) activity can be used to link EEG recordings to epileptic and non-epileptic subjects. These two studies required high computational load and could be addressed thanks to NeuralStudio. From a medical perspective, both methods proved the merits of NeuralStudio in brain research applications. For its outstanding features, NeuralStudio has been recently awarded a patent (US patent No. 7502763).

Glia-augmented artificial neural networks: foundations and applications

Information processing in the human brain has always been considered as a source of inspiration in Artificial Intelligence; in particular, it has led researchers to develop different tools such as artificial neural networks. Recent findings in Neurophysiology provide evidence that not only neurons but also isolated and networks of astrocytes are responsible for processing information in the human brain. Artificial neural net- works (ANNs) model neuron-neuron communications. Artificial neuron-glia networks (ANGN), in addition to neuron-neuron communications, model neuron-astrocyte con- nections. In continuation of the research on ANGNs, first we propose, and evaluate a model of adaptive neuro fuzzy inference systems augmented with artificial astrocytes. Then, we propose a model of ANGNs that captures the communications of astrocytes in the brain; in this model, a network of artificial astrocytes are implemented on top of a typical neural network. The results of the implementation of both networks show that on certain combinations of parameter values specifying astrocytes and their con- nections, the new networks outperform typical neural networks. This research opens a range of possibilities for future work on designing more powerful architectures of artificial neural networks that are based on more realistic models of the human brain.