Topological measurement of deep neural networks using persistent homology

Annals of Mathematics and Artificial Intelligence - The inner representation of deep neural networks (DNNs) is indecipherable, which makes it difficult to tune DNN models, control their training...     

Electrical Breakdown Properties and Space Charge Formation in High Temperature Region in Ultraviolet Ray Irradiated PVC

Polyvinyl chloride (PVC) is the most popular insulating material for electric wiring instruments. However, an exothermic reaction above 150 °C may cause deterioration of the insulating properties of PVC. Therefore, it is important to clarify the heat degradation in PVC, not only to investigate the ignition of electrical wiring products but also to use electrical products safely. It is known that ultraviolet (UV) irradiation causes chemical deterioration of PVC and an increase in its conductivity. Generally, it has been thought that the electrical breakdown properties, electrical conduction, and insulating performance are affected by space charge accumulation in an insulating material. A high temperature pulsed electroacoustic (PEA) system usable up to 250 °C has been developed, and the PEA system can measure the space charge distribution and conduction current in the high temperature range simultaneously. In this investigation, the space charge distribution and conduction current were measured up to electrical breakdown in a non‐UV irradiated sample (normal PVC) and in 353 nm and 253 nm UV‐irradiated PVC samples in the range from room temperature to 200 °C in a DC electric field. In the short wavelength UV irradiated PVC sample (253 nm, 300 h), a deterioration of breakdown strength at 90 °C to 150 °C and negative packet‐like charges were observed at 60 °C and 100 °C, a positive charge accumulated in front of both the anode and cathode above 90 °C, and a higher electric field near the cathode side because the positive charge of the cathode side was greater.     

High-resolution long-array thermal ink jet printhead fabricated by anisotropic wet etching and deep Si RIE

This paper describes the fabrication and characterization of a thermal ink jet (TIJ) printhead suitable for high speed and high-quality printing. The printhead has been fabricated by dicing the bonded wafer, which consists of a bubble generating heater plate and a Si channel plate. The Si channel plate consists of an ink chamber and an ink inlet formed by KOH etching, and a nozzle formed by inductively couple plasma reactive ion etching (ICP RIE). The nozzle formed by RIE has squeezed structures, which contribute to high-energy efficiency of drop ejector and, therefore, successful ejection of small ink drop. The nozzle also has a dome-like structure called channel pit, which contributes to high jetting frequency and high-energy efficiency. These two wafers are directly bonded using electrostatic bonding of full-cured polyimide to Si. The adhesive-less bonding provided an ideal shaped small nozzle orifice. Use of the same material (Si substrate) in heater plate and channel plate enables the fabrication of high precision long printhead because no displacement and delamination occur, which are caused by the difference in thermal expansion coefficient between the plates. With these technologies, we have fabricated a 1" long printhead with 832 nozzles having 800 dots per inch (dpi) resolution and a 4 pl. ink drop volume.     

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.     

Operation of superconducting fault current limiter using vacuum interrupter driven by electromagnetic repulsion force for commutating switch

Using conventional high‐temperature superconducting wire, a model superconducting fault current limiter (SFCL) is made and tested. Solenoid coil using Bi2223 silver sheath wire is so made that inductance is as small as possible and a vacuum interrupter is connected in series to it. A conventional reactor coil is connected in parallel. When the fault current flows in this equipment, superconducting wire is quenched and current is transferred into the parallel coil because of voltage drop of superconducting wire. This large current in parallel coil actuates magnetic repulsion mechanism of vacuum interrupter. Due to opening of vacuum interrupter, the current in superconducting wire is broken. By using this equipment, current flow time in superconducting wire can be easily minimized. On the other hand, the fault current is also easily limited by large reactance of parallel coil. © 2008 Wiley Periodicals, Inc. Electr Eng Jpn, 164(1): 52–61, 2008; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/eej.20315     

Thermal stability of reinforced Nb3Sn composite superconductor under cryocooled conditions

Several types of reinforced Nb₃Sn wires have been developed to prevent reduction of superconducting properties by applying a strong electromagnetic force. To fabricate a cryocooled magnet using those reinforced wires, we experimentally measured the minimum quench energy (MQE) under cryocooled conditions of some reinforced Nb₃Sn wires. As a result, it became clear that thermal stability expressed as MQE was controlled by the temperature margin between the temperature of the operating condition and the transition temperature from superconductivity to normal. Using the FEM analysis, it was realized that the cause of the decline in thermal stability for the reinforced wires was the low thermal conductivity of the reinforced materials.     

Aging time dependence of catastrophic optical damage (COD) failureof a 0.98-μm GaInAs-GaInP strained quantum-well laser

In this paper, we studied the aging time dependence of the catastrophic optical damage (COD) failure of an Al-free uncoated 0.98-μm GaInAs-GaInP strained quantum-well laser with an injection current as a parameter. Based on the stress-strength model, we first investigated experimentally the dependence of the critical power level (CPL) at which COD would take place upon the aging time. Then applying a statistical treatment to this result, we found for the first time that CPL data at each aging time could be considered to distribute according to the Weibull statistics, and the decrease rate of the CPL with the aging time depended very strongly on the injection current. Finally, using the relationship between the decrease rate of the CPL with the aging time and the current, we predicted roughly the time of a COD failure occurrence for both large and small current cases. As a result, we clarified that for our Al-free uncoated 0.98-μm laser, a COD failure became a fatal problem in the case of a large-current (high-power) operation     

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.     

25 Gbit/s selector module using 0.2 mu m GaAs MESFET technology

A high-speed selector module has been developed. It is constructed from a selector IC mounted in a ceramic package, a power supply unit, phase shifters, and coaxial cables. The IC was designed using LSCFL and fabricated with 0.2 mu m gate length GaAs MESFETs. The selector module operated above 25 Gbit/s. It is expected to be applied to high-speed IC measurements.<>     

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.