Fast time delay neural networks

This paper presents a new approach to speed up the operation of time delay neural networks. The entire data are collected together in a long vector and then tested as a one input pattern. The proposed fast time delay neural networks (FTDNNs) use cross correlation in the frequency domain between the tested data and the input weights of neural networks. It is proved mathematically and practically that the number of computation steps required for the presented time delay neural networks is less than that needed by conventional time delay neural networks (CTDNNs). Simulation results using MATLAB confirm the theoretical computations.     

Minimizing fiducial localization error using sphere-based registration in jaw tracking

Some of the jaw tracking methods may be limited in terms of their accuracy or clinical applicability. This article introduces the sphere-based registration method to minimize the fiducial (reference landmark) localization error (FLE) in tracking and coregistration of physical and virtual dental models, to enable an effective clinical analysis of the patient’s masticatory functions. In this method, spheres (registration fiducials) are placed on the corresponding polygonal concavities of the physical and virtual dental models based on the geometrical principle that establishes a unique spatial position for a sphere inside an infinite trihedron. The experiments in this study were implemented using an optical system which tracked active tracking markers connected to the upper and lower dental casts. The accuracy of the tracking workflow was confirmed in vitro, based on comparing virtually calculated interocclusal regions of close proximity against the physical interocclusal impressions. The target registration error of the tracking was estimated based on the leave-one-sphere-out method to be the sum of the error of the sensors, i.e., the FLE was negligible. Moreover, based on a user study, the FLE of the proposed method was confirmed to be 5 and 10 times smaller than the FLE of conventional fiducial selections on the physical and virtual models, respectively. The proposed tracking method is non-invasive and appears to be sufficiently accurate. To conclude, the proposed registration and tracking principles can be extended to track any biomedical and non-biomedical geometries that contain polygonal concavities.     

Robust stability of genetic regulatory networks with distributed delay

This paper investigates robust stability of genetic regulatory networks with distributed delay. Different from other papers, distributed delay is induced. It says that the concentration of macromolecule depends on an integral of the regulatory function of over a specified range of previous time, which is more realistic. Based on Lyapunov stability theory and linear matrix inequality (LMI), sufficient conditions for genetic regulatory networks to be global asymptotic stability and robust stability are derived in terms of LMI. Two numerical examples are given to illustrate the effectiveness of our theoretical results.     

Synchronization of chaotic nonlinear continuous neural networks with time-varying delay

In this paper, the synchronization problem for delayed continuous time nonlinear complex neural networks is considered. The delay dependent state feed back synchronization gain matrix is obtained by considering more general case of time-varying delay. Using Lyapunov stability theory, the sufficient synchronization criteria are derived in terms of Linear Matrix Inequalities (LMIs). By decomposing the delay interval into multiple equidistant subintervals, Lyapunov-Krasovskii functionals (LKFs) are constructed on these intervals. Employing these LKFs, new delay dependent synchronization criteria are proposed in terms of LMIs for two cases with and without derivative of time-varying delay. Numerical examples are illustrated to show the effectiveness of the proposed method.     

A ternary logic model for recurrent neuromime networks with delay

 In contrast to popular recurrent artificial neural network (RANN) models, biological neural networks have unsymmetric structures and incorporate significant delays as a result of axonal propagation. Consequently, biologically inspired neural network models are more accurately described by nonlinear differential-delay equations rather than nonlinear ordinary differential equations (ODEs), and the standard techniques for studying the dynamics of RANNs are wholly inadequate for these models. This paper develops a ternary-logic based method for analyzing these networks. Key to the technique is the realization that a nonzero delay produces a bounded stability region. This result significantly simplifies the construction of sufficient conditions for characterizing the network equilibria. If the network gain is large enough, each equilibrium can be classified as either asymptotically stable or unstable. To illustrate the analysis technique, the swim central pattern generator (CPG) of the sea slug Tritonia diomedea is examined. For wide range of reasonable parameter values, the ternary analysis shows that none of the network equilibria are stable, and thus the network must oscillate. The results show that complex synaptic dynamics are not necessary for pattern generation. Received: 15 June 1994/Accepted in revised form: 10 February 1995     

Robust filtering for stochastic genetic regulatory networks with time-varying delay

This paper addresses the robust filtering problem for a class of linear genetic regulatory networks (GRNs) with stochastic disturbances, parameter uncertainties and time delays. The parameter uncertainties are assumed to reside in a polytopic region, the stochastic disturbance is state-dependent described by a scalar Brownian motion, and the time-varying delays enter into both the translation process and the feedback regulation process. We aim to estimate the true concentrations of mRNA and protein by designing a linear filter such that, for all admissible time delays, stochastic disturbances as well as polytopic uncertainties, the augmented state estimation dynamics is exponentially mean square stable with an expected decay rate. A delay-dependent linear matrix inequality (LMI) approach is first developed to derive sufficient conditions that guarantee the exponential stability of the augmented dynamics, and then the filter gains are parameterized in terms of the solution to a set of LMIs. Note that LMIs can be easily solved by using standard software packages. A simulation example is exploited in order to illustrate the effectiveness of the proposed design procedures.     

Gene networks inference using dynamic Bayesian networks

This article deals with the identification of gene regulatory networks from experimental data using a statistical machine learning approach. A stochastic model of gene interactions capable of handling missing variables is proposed. It can be described as a dynamic Bayesian network particularly well suited to tackle the stochastic nature of gene regulation and gene expression measurement. Parameters of the model are learned through a penalized likelihood maximization implemented through an extended version of EM algorithm. Our approach is tested against experimental data relative to the S.O.S. DNA Repair network of the Escherichia coli bacterium. It appears to be able to extract the main regulations between the genes involved in this network. An added missing variable is found to model the main protein of the network. Good prediction abilities on unlearned data are observed. These first results are very promising: they show the power of the learning algorithm and the ability of the model to capture gene interactions.     

Strain differences in delay discounting using inbred rats

A heightened aversion to delayed rewards is associated with substance abuse and numerous other neuropsychiatric disorders. Many of these disorders are heritable, raising the possibility that delay aversion may also have a significant genetic or heritable component. To examine this possibility, we compared delay discounting in six inbred strains of rats (Brown Norway, Copenhagen, Lewis, Fischer, Noble and Wistar Furth) using the adjusting amount procedure, which provides a measure of the subjective value of delayed rewards. The subjective value of rewards decreased as the delay to receipt increased for all strains. However, a main effect of strain and a strain × delay interaction indicated that some strains were more sensitive to the imposition of delays than others. Fitting a hyperbolic discount equation showed significant strain differences in sensitivity to delay ( k ). These data indicate that there are significant strain differences in delay discounting. All strains strongly preferred the 10% sucrose solution (the reinforcer in the delay discounting task) over water and the amount of sucrose consumed was correlated with sensitivity to delay. Locomotor activity was not correlated with delay discounting behavior. Additional research will be required to disentangle genetic influences from maternal effects and to determine how these factors influence the underlying association between heightened delay discounting and neuropsychiatric disorders.     

Long-term activity-dependent plasticity of action potential propagation delay and amplitude in cortical networks

Background

The precise temporal control of neuronal action potentials is essential for regulating many brain functions. From the viewpoint of a neuron, the specific timings of afferent input from the action potentials of its synaptic partners determines whether or not and when that neuron will fire its own action potential. Tuning such input would provide a powerful mechanism to adjust neuron function and in turn, that of the brain. However, axonal plasticity of action potential timing is counter to conventional notions of stable propagation and to the dominant theories of activity-dependent plasticity focusing on synaptic efficacies.

Methodology/Principal Findings

Here we show the occurrence of activity-dependent plasticity of action potential propagation delays (up to 4 ms or 40% after minutes and 13 ms or 74% after hours) and amplitudes (up to 87%). We used a multi-electrode array to induce, detect, and track changes in propagation in multiple neurons while they adapted to different patterned stimuli in controlled neocortical networks in vitro. The changes did not occur when the same stimulation was repeated while blocking ionotropic gabaergic and glutamatergic receptors. Even though induction of changes in action potential timing and amplitude depended on synaptic transmission, the expression of these changes persisted in the presence of the synaptic receptor blockers.

Conclusions/Significance

We conclude that, along with changes in synaptic efficacy, propagation plasticity provides a cellular mechanism to tune neuronal network function in vitro and potentially learning and memory in the brain.     

Robust exponential stability of markovian jumping neural networks with time-varying delay

This paper considers the robust stability of a class of neural networks with Markovian jumping parameters and time-varying delay. By employing a new Lyapunov-Krasovskii functional, a sufficient condition for the global exponential stability of the delayed Markovian jumping neural networks is established. The proposed condition is also extended to the uncertain cases, which are shown to be the improvement and extension of the existing ones. Finally, the validity of the results are illustrated by an example.     

Measurement of delay discounting using trial-by-trial consequences

DELAY DISCOUNTING IN HUMANS WAS INVESTIGATED USING THREE DIFFERENT PROCEDURES: a frequently used discounting procedure with hypothetical rewards and delays; a procedure with hypothetical rewards and delays compressed down to much smaller values; and a contingent procedure in which each choice had a direct consequence. In the contingent procedure, on every trial, participants actually experienced the delay and obtained the reward amount associated with their choice. Each participant was exposed to all three procedures. Orderly temporal discounting patterns were obtained in all three procedures and described well by a hyperbolic model. Comparisons of the data revealed patterns unique to each procedure. The distributions of the discounting measures differed across the three procedures. In the contingent procedure, several subjects showed no discounting, e.g. complete self-control. Procedural factors in studies of impulsivity are discussed, and suggestions are offered for experiments in which the contingent-discounting procedure may prove useful.     

Maintaining synchronization by decentralized feedback control in time delay neural networks with parameter uncertainties

A decentralized feedback control scheme is proposed to synchronize linearly coupled identical neural networks with time-varying delay and parameter uncertainties. Sufficient condition for synchronization is developed by carefully investigating the uncertain nonlinear synchronization error dynamics in this article. A procedure for designing a decentralized synchronization controller is proposed using linear matrix inequality (LMI) technique. The designed controller can drive the synchronization error to zero and overcome disruption caused by system uncertainty and external disturbance.     

Assessing rbf networks using DELVE

In this paper, different methods for training radial basis function (RBF) networks for regression problems are described and illustrated. Then, using data from the DELVE archive, they are empirically compared with each other and with some other well known methods for machine learning. Each of the RBF methods performs well on at least one DELVE task, but none are as consistent as the best of the other non-RBF methods.     

Adaptive networks using learning matrices

Summary The performance of the Learning Matrix (LM) is suitable for the design of adaptive networks of higher complexity. It has been published, how to connect a LM with a generator of patterns (binary or nonbinary) and a ring-counter to result in an automatic classification of the presented patterns. This paper describes, how to connect two LM's to form an Autonomous Learning Matrix Dipole (ALD) and how to organize it, so that it adapts itself to an environment according to a given evaluation scale. For this purpose, a third type of input (beside e and b), namely h seems to be useful. This h-input controls the rate of adaptation of the LM.Using such ALD's, one may design adaptive structures of even higher complexity, for example with an adaptive internal model.The principle of Learning Matrices has been explained in detail (see e.g. IEEE Transactions on Electronic Computers, Vol. EC-12, No. 6, December, 1963, pp. 846–862). Using such learning matrices (LM), one may build up adaptive networks with rather interesting functions. Perhaps they are interesting for the physiologist and psychologist as well as for the engineer. Let us first recall the most essential details of the LM's.

---

Zusammenfassung Die Funktion der Lernmatrix (LM) erlaubt den Entwurf adaptiver Netzwerke höherer Komplexität. Es wurde an anderer Stelle schon beschrieben, wie eine LM (binär oder nichtbinär) mit einem Generator für Eigenschaftssätze und einem Ringzähler zusammengeschaltet werden kann, um eine selbsttätige Klassifikation der angebotenen Eigenschaftssätze zu bewirken. Im vorliegenden Aufsatz wird erklärt, wie zwei LM so zusammengeschaltet werden können, dacß sich ein Autonomer Lernmatrix-Dipol (ALD) ergibt, und wie dieser zu organisieren ist, daß er sich einer gegebenen Außenwelt nach Maßgabe einer vorgegebenen Werteskala anpaßt. Zu diesem Zweck erweist sich außer den bisher beschriebenen beiden Zugangen zur LM (nämlich e und b) ein dritter sehr zweckmäßig, nämlich h. Dieser h-Eingang beeinflußt die Lerngeschwindigkeit der LM.Unter Verwendung solcher ALD's kann man adaptive Strukturen noch höherer Komplexität aufbauen, beispielsweise solche mit adaptivem innerem Modell.

---

---

Visiting Professor of Electrical Engineering Stanford University.     

Minimizing polymerase biases in metabarcoding

DNA metabarcoding is an increasingly popular method to characterize and quantify biodiversity in environmental samples. Metabarcoding approaches simultaneously amplify a short, variable genomic region, or “barcode,” from a broad taxonomic group via the polymerase chain reaction (PCR), using universal primers that anneal to flanking conserved regions. Results of these experiments are reported as occurrence data, which provide a list of taxa amplified from the sample, or relative abundance data, which measure the relative contribution of each taxon to the overall composition of amplified product. The accuracy of both occurrence and relative abundance estimates can be affected by a variety of biological and technical biases. For example, taxa with larger biomass may be better represented in environmental samples than those with smaller biomass. Here, we explore how polymerase choice, a potential source of technical bias, might influence results in metabarcoding experiments. We compared potential biases of six commercially available polymerases using a combination of mixtures of amplifiable synthetic sequences and real sedimentary DNA extracts. We find that polymerase choice can affect both occurrence and relative abundance estimates and that the main source of this bias appears to be polymerase preference for sequences with specific GC contents. We further recommend an experimental approach for metabarcoding based on results of our synthetic experiments.     

Influence of time delay and channel blocking on multiple coherence resonance in Hodgkin-Huxley neuron networks

Toxins such as tetraethylammonium (TEA) and tetrodotoxin (TTX) may reduce the number of working potassium and sodium ion channels by poisoning and making them blocked, respectively. In this paper, we study how channel blocking (CB) affects the time delay-induced multiple coherence resonance (MCR), i.e., a phenomenon that the spiking of neuronal networks intermittently reaches the most ordered state, in stochastic Hodgkin-Huxley neuron networks. It is found that potassium and sodium CB have distinct effects. For potassium CB, the MCR occurs more frequently as the CB develops, but for sodium CB the MCR is badly impaired and only the first coherence resonance (CR) holds and, consequently, the MCR evolves into a single CR as sodium CB develops. We found for sodium CB the spiking becomes disordered at larger delay lengths, which may be the reason for the destruction of the MCR. The underlying mechanism is briefly discussed in terms of distinct effects of potassium and sodium CB on the spiking activity. These results show that potassium CB can increase the frequency of MCR with time delay, but sodium CB may suppress and even destroy the delay-induced MCR. These findings may help to understand the joint effects of CB and time delay on the spiking coherence of neuronal networks.