Stability and Synchronization for Discrete-Time Complex-Valued Neural Networks with Time-Varying Delays

In this paper, the synchronization problem for a class of discrete-time complex-valued neural networks with time-varying delays is investigated. Compared with the previous work, the time delay and parameters are assumed to be time-varying. By separating the real part and imaginary part, the discrete-time model of complex-valued neural networks is derived. Moreover, by using the complex-valued Lyapunov-Krasovskii functional method and linear matrix inequality as tools, sufficient conditions of the synchronization stability are obtained. In numerical simulation, examples are presented to show the effectiveness of our method.     

Beam forming of ultra-wideband pulses by a complex-valued spatio-temporal multilayer neural network

We present a neuro-beam former of ultra-wideband (UWB) pulses employing complex-valued spatio-temporal multilayer neural network, where complex-valued back propagation through time (CV-BPTT) is used as a learning algorithm. The system performance is evaluated with a UWB mono-cycle pulse. Simulation results in suppressing multiple UWB interferes and in steering to multiple desired UWB pulses, demonstrates the applicability of the proposed system.     

Frequency-multiplexing ability of complex-valued Hebbian learning in logic gates

Lightwave has attractive characteristics such as spatial parallelism, temporal rapidity in signal processing, and frequency band vastness. In particular, the vast carrier frequency bandwidth promises novel information processing. In this paper, we propose a novel optical logic gate that learns multiple functions at frequencies different from one another, and analyze the frequency-domain multiplexing ability in the learning based on complex-valued Hebbian rule. We evaluate the averaged error function values in the learning process and the error probabilities in the realized logic functions. We investigate optimal learning parameters as well as performance dependence on the number of learning iterations and the number of parallel paths per neuron. Results show a trade-off among the learning parameters such as learning time constant and learning gain. We also find that when we prepare 10 optical path differences and conduct 200 learning iterations, the error probability completely decreases to zero in a three-function multiplexing case. However, at the same time, the error probability is tolerant of the path number. That is, even if the path number is reduced by half, error probability is found almost zero. The results can be useful to determine neural parameters for future optical neural network systems and devices that utilize the vast frequency bandwidth for frequency-domain multiplexing.     

Applying a neural network to predict the thermodynamic parameters for an expanded nearest-neighbor model

Predicting the secondary and tertiary structure of RNAs largely depends on our capabilities in estimating the thermodynamics of RNA duplexes. In this work, an expanded nearest-neighbor model, designated INN-48, is established. The thermodynamic parameters of this model are predicted using both multiple linear regression analysis and neural network analysis. It is suggested that due to the increase in the number of parameters and the insufficiency of the existing data, neural network analysis results in more reliable predictions. Furthermore, it is suggested that INN-48 can be used to estimate the thermodynamics of RNA duplex formation for longer sequences, whereas INN-HB, the previous model on which INN-48 is based, can be used for short sequences.     

Temporal classification of Drosophila segmentation gene expression patterns by the multi-valued neural recognition method

In order to reconstruct the establishment of the body pattern over time in Drosophila embryos, we have developed automated methods for detecting the age of an embryo on the basis of knowledge about its gene expression patterns. In this paper we perform temporal classification of confocal images of expression patterns of genes controlling segmentation by means of a neural network based on multi-valued neurons (MVN). MVN are artificial neural processing elements with complex-valued weights and high functionality, which proved to be efficient for solving the image recognition problems. The results obtained by this method confirm its efficiency for image recognition and indicate that the method can detect characteristic features of expression patterns which mark their development over time.     

Coupling isoprene and monoterpene emissions from Amazonian tree species with physiological and environmental parameters using a neural network approach

The ability to predict isoprene emissions from plants is important for predicting atmospheric chemistry. To improve the basis for prediction capability, data obtained from continuous field measurements of isoprene and monoterpene emissions from three Amazonian tree species were related to observed environmental and leaf physiological parameters using a new neural network approach. The environmental parameters included leaf temperature, light, relative humidity, water vapour pressure deficit, and the history of ambient temperature and ozone concentration, whereas the physiological parameters included stomatal conductance, assimilation and intercellular CO₂ concentration. The neural approach with 24 different combinations of these parameters was applied to predict the emission variability observed during short time periods (2–3 d) with individual tree branches and, on a longer-term scale, in aggregated data sets from different seasons, leaf developmental stage, and light environment. The results were compared to the quasi standard emission algorithm for isoprene. On the short-term scale, good agreement (r²≈ 0.9) was obtained between observations and predictions of the standard algorithm as well as predictions of the neural network using the same input parameters (leaf temperature and light). When these predictors were used to model the long-term emission variability, r² was reduced to < 0.5 for both approaches. Remarkably, for the neural technique, more than 50% of the unexplained variance could be explained by the mean temperature of the preceding 36 h. An even better network performance was obtained with physiological parameter combinations (r² > 0.9) suggesting a strong and applicable link between isoprenoid emission and leaf primary metabolism.     

Global asymptotic stability of complex-valued neural networks with additive time-varying delays

In this paper, we extensively study the global asymptotic stability problem of complex-valued neural networks with leakage delay and additive time-varying delays. By constructing a suitable Lyapunov–Krasovskii functional and applying newly developed complex valued integral inequalities, sufficient conditions for the global asymptotic stability of proposed neural networks are established in the form of complex-valued linear matrix inequalities. This linear matrix inequalities are efficiently solved by using standard available numerical packages. Finally, three numerical examples are given to demonstrate the effectiveness of the theoretical results.     

An improved genetic algorithm based fuzzy-tuned neural network

This paper presents a fuzzy-tuned neural network, which is trained by an improved genetic algorithm (GA). The fuzzy-tuned neural network consists of a neural-fuzzy network and a modified neural network. In the modified neural network, a neuron model with two activation functions is used so that the degree of freedom of the network function can be increased. The neural-fuzzy network governs some of the parameters of the neuron model. It will be shown that the performance of the proposed fuzzy-tuned neural network is better than that of the traditional neural network with a similar number of parameters. An improved GA is proposed to train the parameters of the proposed network. Sets of improved genetic operations are presented. The performance of the improved GA will be shown to be better than that of the traditional GA. Some application examples are given to illustrate the merits of the proposed neural network and the improved GA.     

Feedback enhances feedforward figure-ground segmentation by changing firing mode

In the visual cortex, feedback projections are conjectured to be crucial in figure-ground segregation. However, the precise function of feedback herein is unclear. Here we tested a hypothetical model of reentrant feedback. We used a previous developed 2-layered feedforward spiking network that is able to segregate figure from ground and included feedback connections. Our computer model data show that without feedback, neurons respond with regular low-frequency (～9 Hz) bursting to a figure-ground stimulus. After including feedback the firing pattern changed into a regular (tonic) spiking pattern. In this state, we found an extra enhancement of figure responses and a further suppression of background responses resulting in a stronger figure-ground signal. Such push-pull effect was confirmed by comparing the figure-ground responses with the responses to a homogenous texture. We propose that feedback controls figure-ground segregation by influencing the neural firing patterns of feedforward projecting neurons.     

Diagnostic parameters in liquid-based cervical cytology using a coagulant suspension fixative

OBJECTIVE: To evaluate in detail the morphology of cervical cell samples suspended in the coagulant fixative BoonFix (Finetec, Tokyo, Japan) in liquid-based Papspin slides (Thermo Shandon, Pittsburgh, Pennsylvania, U.S.A) to detect shifts in diagnostic parameters for infections and neoplasia. STUDY DESIGN: Split samples of 1,010 cases were collected. All Papspin slides were scanned with neural network technology. In 849 cases the diagnosis was "within normal limits"; in 22 cases it was preneoplasia. In 151 special cases conventional smears were compared with thin-layer slides. RESULTS: In 85% of the 151 special cases, a shift of the diagnostic parameter was observed in the Papspin slide. The parameter adhesion of inflammatory cells to epithelial cells was easier to discern in 94% of the cases, and adhesion of microorganisms varied 43-100%. Koilocytosis was more visible in 79%. Prominent nucleoli in atypical and malignant cells were enhanced in 50-100% of cases with preneoplasia. The fact that the cells on the Papspin slide were no longer present in diagnostic streaks posed a problem only in the case of follicular cervicitis. CONCLUSION: The shifts in parameters facilitated the diagnostic process. BoonFix permits the screening of liquid-based Papspin slides, which have proven to be well suited to automated neural network scanning.     

一种自优化RBF神经网络的叶绿素a浓度时序预测模型

藻类水华发生过程具有复杂性、非线性、时变性等特点,其准确预测一直是一个国际性难题.以天津市于桥水库为研究对象,根据2000年1月至2003年12月常规监测的水生生态数据(采样周期为10 d),提出了一种结合时序方法的可自优化RBF神经网络智能预测模型,对判断藻类水华的重要指标叶绿素a浓度进行预测.研究了训练样本量及RBF神经网络扩展速度SPREAD值的可自优化性能,以及该模型用于于桥水库叶绿素a浓度的短期变化趋势预测的可行性.结果表明,预测性能指标随SPREAD值及样本量不同发生变化,该预测模型能自动寻到最优SPREAD值,并发现至少需要约两年的训练样本量才能达到较好预测效果.当样本量为105,SPREAD值为10时,预测效果最好,精度较高,预测值与实测值的相关系数R达到0.982.该方法对水库的藻类水华预警有一定的参考价值.     

Bayesian Neural Network Models for Censored Data

Neural networks are considered by many to be very promising tools for classification and prediction. The flexibility of the neural network models often result in over-fit. Shrinking the parameters using a penalized likelihood is often used in order to overcome such over-fit. In this paper we extend the approach proposed by FARAGGI and SIMON (1995a) to modeling censored survival data using the input-output relationship associated with a single hidden layer feed-forward neural network. Instead of estimating the neural network parameters using the method of maximum likelihood, we place normal prior distributions on the parameters and make inferences based on derived posterior distributions of the parameters. This Bayesian formulation will result in shrinking the parameters of the neural network model and will reduce the over-fit compared with the maximum likelihood estimators. We illustrate our proposed method on a simulated and a real example.     

Models of Hopfield-type quaternion neural networks and their energy functions

Recently models of neural networks that can directly deal with complex numbers, complex-valued neural networks, have been proposed and several studies on their abilities of information processing have been done. Furthermore models of neural networks that can deal with quaternion numbers, which is the extension of complex numbers, have also been proposed. However they are all multilayer quaternion neural networks. This paper proposes models of fully connected recurrent quaternion neural networks, Hopfield-type quaternion neural networks. Since quaternion numbers are non-commutative on multiplication, some different models can be considered. We investigate dynamics of these proposed models from the point of view of the existence of an energy function and derive their conditions for existence.     

Backpropagation algorithm adaptation parameters using learning automata

Despite of the many successful applications of backpropagation for training multi-layer neural networks, it has many drawbocks. For complex problems it may require a long time to train the networks, and it may not train at all. Long training time can be the result of the non-optimal parameters. It is not easy to choose appropriate value of the parameters for a particular problem. In this paper, by interconnection of fixed structure learning automata (FSLA) to the feedforward neural networks, we apply learning automata (LA) scheme for adjusting these parameters based on the observation of random response of neural networks. The main motivation in using learning automata as an adaptation algorithm is to use its capability of global optimization when dealing with multi-modal surface. The feasibility of proposed method is shown through simulations on three learning problems: exclusive-or, encoding problem, and digit recognition. The simulation results show that the adaptation of these parameters using this method not only increases the convergence rate of learning but it increases the likelihood of escaping from the local minima.     

Maximum Likelihood Neural Network Prediction Models

Neural networks have received much attention in recent years mostly by non-statisticians. The purpose of this paper is to incorporate neural networks in a non-linear regression model and obtain maximum likelihood estimates of the network parameters using a standard Newton-Raphson algorithm. We use maximum likelihood estimators instead of the usual back-propagation technique and compare the neural network predictions with predictions of quadratic regression models and with non-parametric nearest neighbor predictions. These comparisons are made using data generated from a variety of functions. Because of the number of parameters involved, neural network models can easily over-fit the data, hence validation of results is crucial.     

Novel hybrid method for the evaluation of parameters contributing in determination of protein structural classes

Due to the increasing gap between structure-determined and sequenced proteins, prediction of protein structural classes has been an important problem. It is very important to use efficient sequential parameters for developing class predictors because of the close sequence-structure relationship. The multinomial logistic regression model was used for the first time to evaluate the contribution of sequence parameters in determining the protein structural class. An in-house program generated parameters including single amino acid and all dipeptide composition frequencies. Then, the most effective parameters were selected by a multinomial logistic regression. Selected variables in the multinomial logistic model were Valine among single amino acid composition frequencies and Ala-Gly, Cys-Arg, Asp-Cys, Glu-Tyr, Gly-Glu, His-Tyr, Lys-Lys, Leu-Asp, Leu-Arg, Pro-Cys, Gln-Met, Gln-Thr, Ser-Trp, Val-Asn and Trp-Asn among dipeptide composition frequencies. Also a neural network model was constructed and fed by the parameters selected by multinomial logistic regression to build a hybrid predictor. In this study, self-consistency and jackknife tests on a database constructed by Zhou [1998. An intriguing controversy over protein structural class prediction. J. Protein Chem. 17(8), 729-738] containing 498 proteins are used to verify the performance of this hybrid method, and are compared with some of prior works. The results showed that our two-stage hybrid model approach is very promising and may play a complementary role to the existing powerful approaches.     

The Fine Structure of the Macrogametocytes of Adelina tribolii Bhatia, 1937 (Eucoccidia, Telosporea) from the Fat Body of the Beetle Tribolium castaneum Hbst.

SYNOPSIS. Macrogametocytes of the coccidium Adelina tribolii Bhatia, 1937 are described from the time when they settle in the fat body of the host and form periparasitic vacuoles around them to the stage of microgametocyte occurrence and the beginning of syzygy formation.
The macrogametocyte is surrounded by a 2-layered pellicle 50 mμ thick. Its continuity is interrupted by one or several micropores 40 mμ across and 86 mμ deep.
The cytoplasm of the parasite contains numerous vesicles and lamellae of rough and smooth endoplasmic reticulum. Mitochondria of various sizes have short tubules. The macrogametocyte contains a variable number of dark bodies 1.4-2.4 μ in diameter. It also contains several vacuoles up to 1.2 μ which are covered with a 3-layered membrane and enclose a granular material.
In old macrogametocytes in syzygy multivesicular bodies develop which measure up to 2.4 by 1.6 μ. Several smaller vacuoles containing granular material are also a constituent of the electrondense basic substance of these corpuscles.
Paraglycogen granules 1.4 by 0.9 A occur in old macrogametocytes and are situated inside the vacuoles which are not bordered by a membrane. The numbers and size of these granules increase with the age of the parasite. The Golgi complex lies close to the nucleus.
The nucleus, 6-8.5 μ in diameter, is in the center of the macrogametocyte and contains a large eccentric nucleolus. The nuclear membrane is 2-layered and has many pores.     

Intratree Variation in Fruit Production and Implications for Primate Foraging

We tested the hypothesis that fruit quantity and quality vary vertically within trees. We quantified intratree fruit production before exploitation by frugivores at different heights in 89 trees from 17 species fed on by primates in Kibale National Park, Uganda. We also conducted a pilot study to determine if the nutritional value of fruit varied within tree crowns. Depending on the species and crown size, we divided tree canopies into 2 or 3 vertical layers. In 2-layered trees, upper crowns produced fruits that were 9.6–30.1% bigger and 0.52–140 times the densities of those from lower crowns, with one exception. Among 2-layered trees, upper crowns produced a mean of 46.9 fruits/m³ (median 12.1), while lower crowns produced a mean of 14.1 fruits/m³ (median 2.5). Among 3-layered trees, upper crowns produced a mean density of 49.9 fruits/m³ (median 12.5), middle crowns a mean of 16.8 fruits/m³ (median 6.6), and lower crowns a mean of 12.8 fruits/m³ (median 1.8). Dry pulp and moisture were systematically greater per fruit in the highest compared to the lowest canopy layers (22.4% and 16.4% respectively in 2-layered trees, 49.7% and 21.8% respectively in 3-layered trees). In 1 tree of Diospyros abyssinica, a pilot nutritional study showed that upper crown ripe fruit contained 41.9% more sugar, 8.4% more crude proteins, and 1.8 times less of the potentially toxic saponin than lower crown ripe fruit, but the result needs to be verified with more individuals and species of trees. We discuss the consequences of intratree variations in fruit production with respect to competition among frugivorous primates.