Network information criterion-determining the number of hiddenunits for an artificial neural network model

The problem of model selection, or determination of the number of hidden units, can be approached statistically, by generalizing Akaike's information criterion (AIC) to be applicable to unfaithful (i.e., unrealizable) models with general loss criteria including regularization terms. The relation between the training error and the generalization error is studied in terms of the number of the training examples and the complexity of a network which reduces to the number of parameters in the ordinary statistical theory of AIC. This relation leads to a new network information criterion which is useful for selecting the optimal network model based on a given training set.     

DNA damage induced by m-phenylenediamine and its derivative in the presence of copper ion

We have previously reported that long-term priming of human polymorphonuclear neutrophilic granulocytes (PMN) with interferon-gamma (IFN-gamma) increased the fMLP-stimulated calcium influx. We now show that also after short-term incubation with IFN-gamma, PMN calcium metabolism is modulated. Single adherent cells in three different calcium-containing buffers (high, normal, and low [Ca2+]) were stimulated with the bacterial peptide fMLP or the Ca-ATPase inhibitor thapsigargin (Tg) after about 5 min preincubation with IFN-gamma. The results of this protocol indicated that IFN-gamma increases both calcium influx and calcium sequestration. Store dependent Ca2+ influx, directly measured on readdition of calcium to Tg-treated cells incubated in EGTA buffer, was significantly enhanced in IFN-gamma-treated cells. This effect of IFN-gamma was enhanced by the tyrosine kinase inhibitor herbimycin A. Strikingly, in low extracellular calcium concentrations, IFN-gamma induced calcium transients in 20%-60% of the cells. The proportion of PMN responding with Ca2+ transients increased with decreasing extracellular calcium concentration. Average lagtime from addition of IFN-gamma to a response that could be measured was 7.3 sec, and average increase in [Ca2+] above the basal level was 790 nM. These IFN-gamma-induced transients could not be depressed by herbimycin A. Thus, IFN-gamma can increase capacitative calcium influx, induce calcium transients, and possibly affect calcium sequestration in human PMN.     

A Composite Index of Compliance for Chronic In‐Center Hemodialysis Patients

We developed a composite compliance index as the sum of the compliance scores for interdialytic weight gain (IDWG), pre‐dialysis serum potassium and phosphorus concentrations (each scored from zero to 3, with 3 indicating the poorest compliance), and skipping hemodialysis sessions (scored from zero to 9, with 9 indicating the poorest compliance). We used this composite score to prospectively evaluate compliance in 25 prevalent hemodialysis patients over a period of 1 year. We then followed these patients for another 3.5 years. The patients studied were divided into two groups: group A (poor compliance) consisted of 9 subjects with composite score ≥ 9 (13.2 ± 3.2); group B (better compliance) consisted of 16 subjects with composite score < 9 (4.7 ± 1.8). Age, duration of hemodialysis, and frequency of diabetes mellitus did not differ between the groups. Group A contained higher fractions of subjects with history of alcoholism (66.7% vs 12.5%, p = 0.010), other substance addiction (44.4% vs 0%, p = 0.010), and severe psychosocial problems (88.9% vs 18.8%, p = 0.002). Mean survival from the beginning of observation, estimated by actuarial life‐table survival analysis, was 1.19 years in group A and 2.60 years in group B (p = 0.0265). A composite compliance index incorporating domains indicating adherence to diet, medications, and dialysis schedule identified other behavioral problems in poorly compliant patients. Hemodialysis patients characterized by this composite index as poorly compliant had shortened survival.     

Survey of switching techniques in high-speed networks and their performance

One of the most promising approaches for high speed networks for integrated service applications is fast packet switching, or ATM (asynchronous transfer mode). ATM can be characterized by very high speed transmission links and simple, hard-wired protocols within a network. To match the transmission speed of the network links, and to minimize the overhead due to the processing of network protocols, the switching of cells is done in hardware switching fabrics in ATM networks. A number of designs have been proposed for implementing ATM switches. Although many differences exist among the proposals, the vast majority of them are based on self-routeing multistage interconnection networks. This is because of the desirable features of multi-stage interconnection networks such as self-routeing capability and suitability for VLSI implementation. Existing ATM switch architectures can be classified into two major classes: blocking switches, where blockings of cells may occur within a switch when more than one cell contends for the same internal link, and non-blocking switches, where no internal blocking occurs. A large number of techniques have also been proposed to improve the performance of blocking and non-blocking switches. In this paper, we present an extensive survey of the existing proposals for ATM switch architectures, focusing on their performance issues.     

H∞ control versus disturbance-observer-basedcontrol

In the motion control field, a disturbance observer-based disturbance canceling control is often used as a robust control methodology. However, this method is nothing more than an alternative design of an integral controller, and the robust stability issue cannot be directly accounted for. In this paper, an extended H_∞ control scheme is proposed as a new robust motion control method which achieves the disturbance cancellation ability and guarantees robust stability automatically     

Visualizing the structure of Web communities based on data acquired from a search engine

Discovery of Web communities, groups of Web pages sharing common interests, is important for assisting users' information retrieval from the Web. This paper describes a method for visualizing Web communities and their internal structures. visualization of Web communities in the form of graphs enables users to access related pages easily, and it often reflects the characteristics of the Web communities. Since related Web pages are often co-referred from the same Web page, the number of co-occurrences of references in a search engine is used for measuring the relation among pages. Two URLs are given to a search engine as keywords, and the value of the number of pages searched from both URLs divided by the number of pages searched from either URL, which is called the Jaccard coefficient, is calculated as the criteria for evaluating the relation between the two URLs. The value is used for determining the length of an edge in a graph so that vertices of related pages will be located close to each other. Our visualization system based on the method succeeds in clarifying various genres of Web communities, although the system does not interpret the contents of the pages. The method of calculating the Jaccard coefficient is easily processed by computer systems, and it is suitable for visualization using the data acquired from a search engine.     

Young’s modulus of crystalline C60 nanotubes studied by in situ transmission electron microscopy

Bending tests of crystalline nanotubes composed of fullerene C₆₀ molecules are performed inside a high-resolution transmission electron microscope. We fixed one side of a C₆₀ nanotube with a body-centered tetragonal structure with typical inner and outer diameters, i.e., 180 nm and 510 nm, respectively, and then applied concentrated forces on the other side using piezomanipulation of a silicon nanotip. The bending process was observed in situ by transmission electron microscopy with simultaneous measurements of the forces by an optical deflection method. It was found that the Young’s modulus of the nanotube was estimated to be 62–107 GPa, which was 1.1–3.3 times larger than that of C₆₀ nanowhiskers. The result concerning the increase in the Young’s modulus of the C₆₀ nanotube provided an experimental evidence for the structural model composed of an inner core and a surface shell for C₆₀ nanowhiskers.     

Virtual Fiber Configuration for Dynamic Lightpath Establishment in Large-scaled Optical Networks

Recently, progress has been made in the Generalized Multi-Protocol Label Switching (GMPLS) and Automatic Switched Optical Networks (ASON) standardizations. These technologies realize construction of large-scaled optical networks, interconnections among single-domain Wavelength Division Multiplexing (WDM) networks, and direct communication over multi-domain WDM networks. Meanwhile, it is known that the topology of the Internet exhibits the power-law attribute. Since the topology of the Internet, which is constructed by interconnecting ASs, exhibits the power-law, there is a possibility that large-scale WDM networks, which are constructed by interconnecting WDM networks, will also exhibit the power-law attribute. One of the structural properties of a topology that adheres to the power-law is that most nodes have just a few links, although some have a tremendous number of them. Another property is that the average distance between nodes is smaller than in a mesh-like network. A natural question is how such a structural property performs in WDM networks. In this paper, we first investigate the property of the power-law attribute of physical topologies for WDM networks. We compare the performance of WDM networks with mesh-like and power-law topologies, and show that links connected to high-degree nodes are bottlenecks in power-law topologies. To relax this, we introduce a concept of virtual fiber, which consists of two or more fibers, and propose its configuration method to utilize wavelength resources more effectively. We compare performances of power-law networks with and without our method by computer simulations. The results show that our method reduces the blocking probabilities by more than one order of magnitude.