Parameter Estimation in Groundwater Hydrology Using Artificial Neural Networks

The capability of artificial neural networks to act as universal function approximators has been traditionally used to model problems in which the relation between dependent and independent variables is poorly understood. In this paper, the capability of an artificial neural network to provide a data-driven approximation of the explicit relation between transmissivity and hydraulic head as described by the groundwater flow equation is demonstrated. Techniques are applied to determine the optimal number of nodes and training patterns needed for a neural network to approximate groundwater parameters for a simulated groundwater modeling case study. Furthermore, the paper explains how such an approximation can be used for the purpose of parameter estimation in groundwater hydrology.     

New Approach to Designing Multilayer Feedforward Neural Network Architecture for Modeling Nonlinear Restoring Forces. I: Formulation

This paper addresses the modeling problem of nonlinear and hysteretic dynamic behaviors through a constructive modeling approach which exploits existing mathematical concepts in artificial neural network modeling. In contrast with many neural network applications, which often result in large and complex “black-box” models, here, the writers strive to produce phenomenologically accurate model behavior starting with network architecture of manageable/small sizes. This affords the potential of creating relationships between model parameter values and observed phenomenological behaviors. Here a linear sum of basis functions is used in modeling nonlinear hysteretic restoring forces. In particular, nonlinear sigmoidal activation functions are chosen as the core building block for their robustness in approximating arbitrary functions. The appropriateness and effectiveness of this set of basis function in modeling a wide variety of nonlinear dynamic behaviors observed in structural mechanics are depicted from an algebraic and geometric perspective.     

Feed forward neural networks for the analysis of censored survival data: a partial logistic regression approach

Flexible modelling in survival analysis can be useful both for exploratory and predictive purposes. Feed forward neural networks were recently considered for flexible non-linear modelling of censored survival data through the generalization of both discrete and continuous time models. We show that by treating the time interval as an input variable in a standard feed forward network with logistic activation and entropy error function, it is possible to estimate smoothed discrete hazards as conditional probabilities of failure. We considered an easily implementable approach with a fast selection criteria of the best configurations. Examples on data sets from two clinical trials are provided. The proposed artificial neural network (ANN) approach can be applied for the estimation of the functional relationships between covariates and time in survival data to improve model predictivity in the presence of complex prognostic relationships.     

Peptide design aided by neural networks: biological activity of artificial signal peptidase I cleavage sites

De novo designed signal peptidase I cleavage sites were tested for their biological activity in vivo in an Escherichia coli expression and secretion system. The artificial cleavage site sequences were generated by two different computer-based design techniques, a simple statistical method, and a neural network approach. In previous experiments, a neural network was used for feature extraction from a set of known signal peptidase I cleavage sites and served as the fitness function in an evolutionary design cycle leading to idealized cleavage site sequences. The cleavage sites proposed by the two algorithms were active in vivo as predicted. There seems to be an interdependence between several cleavage site features for the constitution of sequences recognized by signal peptidase. It is concluded that neural networks are useful tools for sequence-oriented peptide design.     

Peptide design by artificial neural networks and computer-based evolutionary search

A technique for systematic peptide variation by a combination of rational and evolutionary approaches is presented. The design scheme consists of five consecutive steps: (i) identification of a "seed peptide" with a desired activity, (ii) generation of variants selected from a physicochemical space around the seed peptide, (iii) synthesis and testing of this biased library, (iv) modeling of a quantitative sequence-activity relationship by an artificial neural network, and (v) de novo design by a computer-based evolutionary search in sequence space using the trained neural network as the fitness function. This strategy was successfully applied to the identification of novel peptides that fully prevent the positive chronotropic effect of anti-beta1-adrenoreceptor autoantibodies from the serum of patients with dilated cardiomyopathy. The seed peptide, comprising 10 residues, was derived by epitope mapping from an extracellular loop of human beta1-adrenoreceptor. A set of 90 peptides was synthesized and tested to provide training data for neural network development. De novo design revealed peptides with desired activities that do not match the seed peptide sequence. These results demonstrate that computer-based evolutionary searches can generate novel peptides with substantial biological activity.     

Barrels VI: Proceedings of a satellite symposium of the 1994 Society for Neuroscience Meeting

Recent neurophysiological studies have revealed the patterns of neuronal activity during the acquisition of goal-directed behaviors, both in single cells, and in large populations of neurons. We propose a model which helps three sets of experimental results in the monkey to be understood: (1) activity of single cells vary greatly and only population activities are causally related to behavior. The model shows how a population of stochastic neurons, whose behaviors vary widely, can learn a skilled conditioned movement with only local activity-dependent synaptic changes. (2) typical changes in neuronal activity occur when the rules governing the behavior are changed, i.e. when the relationship between cues and actions to reach a goal changes over time. There are two types of neuronal patterns during changes in reward contingency: a monotonic increasing pattern and a non-monotonic pattern which follows the change in the way the reward is obtained. Units in the model display these two types of change, which correspond to synaptic modifications related to the encoding of the behavioral significance of sensory and motor events. (3) These two patterns of neuronal activity define two populations whose anatomical distributions in the frontal lobe overlap with a gradient organized in the rostro-caudal direction. The model consists of two artificial neural networks, defined by the same set of equations, but which differ in the values of two parameters (P and Q). P defines the adaptive properties of processing units and Q describes the coding of information. The model suggests that a balance in the relative strengths of these parameters distributed along a rostro-caudal gradient can explain the distribution of neuronal types in the frontal lobe of the monkey.     

A neuronal network for computing population vectors in the leech

The correlation of neuronal activity with sensory input and behavioural output has revealed that information is often encoded in the activity of many neurons across a population, that is, a neural population code is used. The possible algorithms that downstream networks use to read out this population code have been studied by manipulating the activity of a few neurons in a population. We have used this approach to study population coding in a small network underlying the leech local bend, a body bend directed away from a touch stimulus. Because of the small size of this network we are able to monitor and manipulate the complete set of sensory inputs to the network. We show here that the population vector formed by the spike counts of the active mechanosensory neurons is well correlated with bend direction. A model based on the known connectivity of the identified neurons in the local bend network can account for our experimental results, and is suitable for reading out the neural population vector. Thus, for the first time to our knowledge, it is possible to link a proposed algorithm for neural population coding with synaptic and network mechanisms in an experimental system.     

beta-endorphin1-31 in the rat pituitary

The behavioral and neural correlates of processing of motor directional information are described for two visuomotor tasks: mental rotation and context-recall. Psychological studies with human subjects suggested that these two tasks involve different time-consuming processes of directional information. Analyses of the activity of single cells and neuronal populations in the motor cortex of behaving monkeys performing in the same tasks provided direct insight into the neural mechanisms involved and confirmed their different nature. In the mental rotation task the patterns of neuronal activity revealed a rotation of the intended direction of movement. In contrast, in the context-recall task the patterns of neural activity identified a switching process of the intended direction of movement.     

Neural feature abstraction from judgments of similarity

The common neural network modeling practice of representing the elements of a task domain in terms of a set of features lacks justification if the features are derived through some form of ad hoc preabstraction. By examining a featural similarity model related to established multidimensional scaling techniques, a neural network is developed that generates features from similarity data and attaches weights to these features. The network performs a constrained search of a continuous solution space to determine the features and uses a previously developed regularization technique to minimize the number of features it derives. The network is demonstrated on artificial data, from which it recovers known features and weights, and on two real data sets involving the similarity of a set of geometric shapes and the abstract conceptual similarities of the 10 Arabic numerals. On the basis of these results, the relationship between the multidimensional scaling approach adopted by the network and an alternative additive clustering approach to feature extraction is discussed.     

Preferential use of the landmark navigational system by head direction cells in rats.

Previous studies have identified a population of cells recorded in the postsubiculum and the anterior thalamic nucleus (ATN) that discharge as a function of an animal's head direction (HD) in the horizontal plane. The present experiments monitored HD cell activity when rats were confronted with a situation in which directional information from internal sensory sources (e.g., proprioceptive, vestibular, or motor efference copy) conflicted with directional information derived from familiar, external landmarks. Results showed that when a salient, familiar cue was reintroduced to rat's environment into a position that conflicted with the cell's current firing direction, HD cells in both the ATN and the postsubiculum shifted their preferred direction to reflect their originally established orientation with this cue. This finding suggests that sensory inputs onto HD cells from external landmark cues are capable of overriding spatial information developed through internal sensory cues. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Estimation of spatiotemporal neural activity using radial basis function networks

We report a method using radial basis function (RBF) networks to estimate the time evolution of population activity in topologically organized neural structures from single-neuron recordings. This is an important problem in neuroscience research, as such estimates may provide insights into systems-level function of these structures. Since single-unit neural data tends to be unevenly sampled and highly variable under similar behavioral conditions, obtaining such estimates is a difficult task. In particular, a class of cells in the superior colliculus called buildup neurons can have very narrow regions of saccade vectors for which they discharge at high rates but very large surround regions over which they discharge at low, but not zero, levels. Estimating the dynamic movement fields for these cells for two spatial dimensions at closely spaced timed intervals is a difficult problem, and no general method has been described that can be applied to all buildup cells. Estimation of individual collicular cells' spatiotemporal movement fields is a prerequisite for obtaining reliable two-dimensional estimates of the population activity on the collicular motor map during saccades. Therefore, we have developed several computational-geometry-based algorithms that regularize the data before computing a surface estimation using RBF networks. The method is then expanded to the problem of estimating simultaneous spatiotemporal activity occurring across the superior colliculus during a single movement (the inverse problem). In principle, this methodology could be applied to any neural structure with a regular, two-dimensional organization, provided a sufficient spatial distribution of sampled neurons is available.     

Mechanisms of developmental regression in autism and the broader phenotype: A neural network modeling approach.

Loss of previously established behaviors in early childhood constitutes a markedly atypical developmental trajectory. It is found almost uniquely in autism and its cause is currently unknown (Baird et al., 2008). We present an artificial neural network model of developmental regression, exploring the hypothesis that regression is caused by overaggressive synaptic pruning and identifying the mechanisms involved. We used a novel population-modeling technique to investigate developmental deficits, in which both neurocomputational parameters and the learning environment were varied across a large number of simulated individuals. Regression was generated by the atypical setting of a single pruning-related parameter. We observed a probabilistic relationship between the atypical pruning parameter and the presence of regression, as well as variability in the onset, severity, behavioral specificity, and recovery from regression. Other neurocomputational parameters that varied across the population modulated the risk that an individual would show regression. We considered a further hypothesis that behavioral regression may index an underlying anomaly characterizing the broader autism phenotype. If this is the case, we show how the model also accounts for several additional findings: shared gene variants between autism and language impairment (Vernes et al., 2008); larger brain size in autism but only in early development (Redcay & Courchesne, 2005); and the possibility of quasi-autism, caused by extreme environmental deprivation (Rutter et al., 1999). We make a novel prediction that the earliest developmental symptoms in the emergence of autism should be sensory and motor rather than social and review empirical data offering preliminary support for this prediction. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

预防中厚板轧制头部下扣途径探讨

中厚板热轧过程中的头部弯曲是一个极其复杂的多元非线函数关系。通过 机上模拟及生产数据的人工神经网络分析,找出了关键因素与弯曲方向、程度的相关关系,从而对因变量实施有效的控制与调节,解决了改造的新轧机轧件头上扣难题。     

An intrinsic source of behavioral regulation that influences discrete responses to cues important for the initiation of suckling.

The present experiment investigated the relationship between motor activity and oral grasping of an artificial nipple in newborn rats. Pups orally grasped the artificial nipple, and they performed more and longer oral grasps in the latter portion of the nipple presentation. Motor activity was cyclical, and this cyclicity was evident before and during presentation of the artificial nipple. The onset of an oral grasp response was preceded by a period of relatively low motor activity, and the termination of a grasp was followed by relatively high motor activity. The newborn rat pup's intrinsic oscillations in motor activity may regulate the expression of discrete responses to cues important for the initiation of suckling. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A Deformation Mechanism Map for the 1.23Cr-1.2Mo-0.26V Rotor Steel and Its Verification Using Neural Networks

A deformation mechanism map is constructed for the 1.23Cr-1.2Mo-0.26V rotor steel as a function of temperature, stress, and strain rate using published creep test results and the current understanding of time dependent deformation mechanisms operative in complex engineering alloys. Instead of diffusional creep, grain boundary sliding (GBS) accommodated by different deformation processes is considered dominant at lower strain rates. The GBS dominated region is further sub-divided into two parts, where GBS is accommodated by wedge type cracking at temperatures below 0.5T/T _m and the accommodation process changes to creep cavitation at temperatures above 0.5T/T _m. The map is verified using experimental data and artificial neural network modeling. The proposed artificial neural network model is capable of predicting the dominance of different deformation mechanisms in 1.23Cr-1.2Mo-0.26V steel over a wide range of stress and temperature. This modeling procedure can potentially be used to construct or expand deformation mechanism maps for other engineering alloys.     

Intelligent Multivariable Modeling of Blast Furnace Molten Iron Quality Based on Dynamic AGA-ANN and PCA

Blast furnace(BF)ironmaking process has complex and nonlinear dynamic characteristics.The molten iron temperature(MIT)as well as Si,P and S contents of molten iron is difficult to be directly measured online,and large-time delay exists in offline analysis through laboratory sampling.A nonlinear multivariate intelligent modeling method was proposed for molten iron quality(MIQ)based on principal component analysis(PCA)and dynamic genetic neural network.The modeling method used the practical data processed by PCA dimension reduction as inputs of the dynamic artificial neural network(ANN).A dynamic feedback link was introduced to produce a dynamic neural network on the basis of traditional back propagation ANN.The proposed model improved the dynamic adaptability of networks and solved the strong fluctuation and resistance problem in a nonlinear dynamic system.Moreover,a new hybrid training method was presented where adaptive genetic algorithms(AGA)and ANN were integrated,which could improve network convergence speed and avoid network into local minima.The proposed method made it easier for operators to understand the inside status of blast furnace and offered real-time and reliable feedback information for realizing close-loop control for MIQ.Industrial experiments were made through the proposed model based on data collected from a practical steel company.The accuracy could meet the requirements of actual operation.     

Neural dynamics of short and medium-term motor control effects of levodopa therapy in Parkinson's disease

A neural network model of movement control in normal and Parkinson's disease (PD) conditions is proposed to simulate the time-varying dose-response relationship underlying the effects of levodopa on movement amplitude and movement duration in PD patients. Short and long-term dynamics of cell activations and neurotransmitter mechanisms underlying the differential expression of neuropeptide messenger RNA within the basal ganglia striatum are modeled to provide a mechanistic account for the effects of levodopa medication on motor performance (e.g. the pharmacodynamics). Experimental and neural network simulation data suggest that levodopa therapy in Parkinson's disease has differential effects on cell activities, striatal neuropeptides, and motor behavior. In particular, it is shown how dopamine depletion in the striatum may modulate differentially the level of substance P and enkephalin messenger RNA in the direct and indirect basal ganglia pathways. This dissociation in the magnitude and timing of peptide expression causes an imbalance in the opponently organized basal ganglia pathways which results in Parkinsonian motor deficits. The model is validated with experimental data obtained from handwriting movements performed by PD subjects before and after medication intake. The results suggest that fine motor control analysis and network modeling of the effects of dopamine in motor control are useful tools in drug development and in the optimization of pharmacological therapy in PD patients.     

Action potential classifiers: a functional comparison of template matching, principal components analysis and an artificial neural network

Multiunit neural activity occurs often in electrophysiological studies when utilizing extracellular electrodes. In order to estimate the activity of the individual neurons each action potential in the recording must be classified to its neuron of origin. This paper compares the accuracy of two traditional methods of action potential classification--template matching and principal components--against the performance of an artificial neural network (ANN). Both traditional methods use averages of action potential shapes to form their corresponding classifiers while the artificial neural network 'learns' a nonlinear relationship between a set of prototype action potentials and assigned classes. The set of prototypic action potentials and the assigned classes is termed the training set. The training set contained action potentials from each class which exhibited the full range of amplitude variability. The ANN provided better classification results and was more robust in analysis of across-animal data sets than either of the traditional action potential classification methods.     

Changing directions of forthcoming arm movements: neuronal activity in the presupplementary and supplementary motor area of monkey cerebral cortex

1. To understand roles played by two cortical motor areas, the presupplementary motor area (pre-SMA) and supplementary motor area (SMA), in changing planned movements voluntarily, cellular activity was examined in two monkeys (Macaca fuscata) trained to perform an arm-reaching task in which they were asked to press one of two target buttons (right or left) in three different task modes. 2. In the first mode (visual), monkeys were visually instructed to result and press either a right or left key in response to a forth coming trigger signal. In the second mode (stay), monkeys were required to wait for the trigger signal and press the same target key as pressed in preceding trials. In the third mode (shift), a 50 Hz auditory cue instructed the monkey to shift the target of the future reach from the previous target to the previous nontarget. 3. While the monkeys were performing this task, we recorded 399 task-related cellular activities from the SMA and the pre-SMA. Among them, we found a group of neurons that exhibited activity changes related specifically to shift trials (shift-related cells). The following properties characterized these 112 neurons. First, they exhibited activity changes after the onset of the 50-Hz auditory cue and before the movement execution when the monkeys were required to change the direction of forthcoming movement. Second, they were not active when the monkeys pressed the same key without changing the direction of the movements. Third, they were not active when the monkeys received the 50-Hz auditory cue but failed to change the direction of the movements by mistake. These observations indicate that the activity of shift-related cells is related to the redirection of the forthcoming movements, but not to the auditory instruction itself or to the location of the target key or the direction of the forthcoming movements. 4. Although infrequently, monkeys made errors in the stay trials and changed directions of the reach voluntarily. In that case, a considerably high proportion of shift-related neurons (12 of 19) exhibited significant activity changes long before initiation of the reach movement. These long-lasting activities were not observed during the preparatory period in correct stay trials, but resembled the shift-related activity observed when the target shift was made toward the same direction. Thus these activity changes were considered to be also related to the process of changing the intended movements voluntarily. 5. We found another population of neurons that showed activity modulation when the target shift was induced by the visual instruction in visual trials (visually guided shift-related neurons). These neurons were active when the light-emitting diode (LED) guided the forthcoming reach to the previous nontarget but not to the previous target. Therefore their activity was not a simple visual response to the LED per se. A majority of them also showed shift-related activity in shift trials (19 of 22 in monkey 2). 6. Neurons exhibiting the shift-related activity were distributed differentially among the two areas. In the pre-SMA, 31% of the neurons recorded showed the shift-related activity, whereas in the SMA, only 7% showed such an activity. These results suggest that pre-SMA and SMA play differential roles in updating the motor plans in accordance with current requirements.     

Classical conditioning in the fetal rat: Reinforcing properties of dynorphin A (1–13).

This study examined the reinforcing properties of dynorphin A (1-13) in a single-trial classical conditioning paradigm in the E20 rat fetus. Injection of dynorphin into the cisterna magna increased fetal motor activity and reduced facial wiping in a test of perioral cutaneous responsiveness. Dynorphin was effective as an unconditioned stimulus (US) in a classical conditioning paradigm using an artificial nipple conditioned stimulus (CS) and dynorphin A (1-13) US. The association between CS and US was dependent on activity in the kappa opioid system. Re-exposure to the artificial nipple CS after a single pairing of the nipple with dynorphin resulted in conditioned activation of the kappa opioid system. Dynorphin A (1-13) functions as a reinforcer for classical conditioning in the rat fetus after intracisternal or intrahemispheric injection with the conditioned response depending on route of administration and site of injection. (PsycINFO Database Record (c) 2010 APA, all rights reserved)