An optimization algorithm based on stochastic sensitivity analysis for noisy objective landscapes

A function minimization algorithm that updates solutions based on approximated derivative information is proposed. The algorithm generates sample points with Gaussian white noise, and approximates derivatives based on stochastic sensitivity analysis. Unlike standard trust region methods which calculate gradients with n or more sample points, where n is the number of variables, the proposed algorithm allows the number of sample points M to be less than n. Furthermore, it ignores small amounts of noise within a trust region. This paper addresses the following two questions: how does the derivative approximation become worse when the number of sample points is small? Can the algorithm find good solutions with inexact derivative information when the objective landscape is noisy? Through intensive numerical experiments using quadratic functions, the algorithm is shown to be able to approximate derivatives when M is about n/10 or more. The experiments using a formulation of the traveling salesman problem show that the algorithm can find reasonably good solutions for noisy objective landscapes with inexact derivatives information.     

Analysis of XANES for identification of highly dispersed transition metal oxides on supports

XANES of vanadium and niobium oxide on silica or alumina have been analyzed quantitatively by a deconvolution technique. Based on the results for reference compounds, local structures of supported vanadium and niobium species were identified. The composition was estimated from difference spectra for the samples which consisted of two kinds of species.     

Wasserabsorptionseigenschaften natürlicher polyionenkomplexe

As the cationic component of natural polymer polyionic complexes, polysaccharide chitosan or polypeptide polylysine, and as the anionic component, polysaccharide carboxymethyl cellulose (=CMC) or polypeptide poly(glutamic acid) (PGA) were used. The degree of chitosan/CMC or polylysine/PGA complex film swelling was minimum, when the amounts of the cationic and anionic groups in these polymers were balanced at 1 : 1 ratio. On the other hand, the degree of chitosan/PGA film swelling was minimum at the ratio with excess of PGA which is a linear flexible chain. Comparing the enthalpies estimated by DSC analysis, it was suggested that the larger quantity of water in the form of nonfreezable bound water was absorbed in the complex films with polysaccharide which is a chain of ring structure. Further, the compressive moduli of swollen complex films were high, when the complex consisted of polysaccharide components.     

Preparation of Copper-Ruby Glasses by Sputtering: The Effect of Atmosphere on the Growth of Copper Particles

Copper-ruby glasses were prepared by the sputtering method and the effect of reducing treatment was examined. The reducing was carried out either during sputtering or heat treatment. Optical absorption was related to the growth of copper particles. The results are summarized as follows: (i) the volume fraction of copper particles in the glasses which are heat-treated in air increases with the addition of hydrogen in the sputtering gas. (ii) The copper particles do not grow larger than 8 nm by heat treatment in reducing atmosphere, whereas they grow as large as 12 nm in air. (iii) Reducing during heat treatment gives a pronounced effect only when sputtering was not carried out under reducing conditions. The origin of these phenomena was also discussed, considering the mechanism of nucleation and growth of copper particles.     

Preparation of zirconia-toughened bioactive glass-ceramics

Bioactive glass-ceramics toughened by tetragonal zirconia polycrystal (TZP) were prepared by hot-pressing mixed powders of the MgO-CaO-P₂O₅-SiO₂ glass and TZP containing 20 to 80% alumina. The bending strength and the fracture toughness of the composite materials were improved compared with those of the material without TZP. These composites showed high bending strengths (400 to 500MPa) and high fracture toughness ( 2.8MPa m^1/2). The existence of a crack deflection mechanism was observed by scanning electron microscopy. After soaking in simulated physiological solution at 100 °C, no phase transformation from tetragonal to monoclinic of TZP in the composites and no degradation in bending strength occurred.     

Enhancement of superconductivity in Pb-Bi-Sn and Pb-Bi-In alloy filaments produced by glass-coated melt spinning

The enhancement ofT _c in Pb-Bi-Sn and Pb-Bi-In system alloy filaments produced by glass-coated melt spinning was investigated as a means of producing a new type of superconducting filament with highT _c. Long filaments of Pb-Bi-Sn alloy withT _c higher than 10 K and Pb-Bi-In alloy withT _c higher than 9 K were obtained from the molten state at a temperature of 1500 K with a winding speed of 2.63 m sec^–1. For example, a Pb₄₅Bi₃₅Sn₂₀ filament withT _c of 10.1 K was 15 m in diameter and polycrystalline with a grain size of 100 nm. The structure of the filament was a mixture of , tin and bismuth phases and a metastable phase of mixed structure of bismuth and supersaturated solid solution of tin in -phase was detected. TheT _c of the filament decreased on heat treatment. A metastable phase of mixed structure of bismuth and -phase was also detected for a Pb₄₅Bi₄₅In₁₀ filament withT _c of 9.3 K. As the metastable phase for the Pb-Bi-In filament was more unstable than that for the Pb-Bi-Sn filament, theT _c of the filament was drastically decreased by heat treatment. The metastable phase was considered to play an important role in the enhancement ofT _c for Pb-Bi-Sn and Pb-Bi-In alloy filaments.     

Links between Immune Cells from the Periphery and the Brain in the Pathogenesis of Epilepsy: A Narrative Review

Accumulating evidence has demonstrated that the pathogenesis of epilepsy is linked to neuroinflammation and cerebrovascular dysfunction. Peripheral immune cell invasion into the brain, along with these responses, is implicitly involved in epilepsy. This review explored the current literature on the association between the peripheral and central nervous systems in the pathogenesis of epilepsy, and highlights novel research directions for therapeutic interventions targeting these reactions. Previous experimental and human studies have demonstrated the activation of the innate and adaptive immune responses in the brain. The time required for monocytes (responsible for innate immunity) and T cells (involved in acquired immunity) to invade the central nervous system after a seizure varies. Moreover, the time between the leakage associated with blood–brain barrier (BBB) failure and the infiltration of these cells varies. This suggests that cell infiltration is not merely a secondary disruptive event associated with BBB failure, but also a non-disruptive event facilitated by various mediators produced by the neurovascular unit consisting of neurons, perivascular astrocytes, microglia, pericytes, and endothelial cells. Moreover, genetic manipulation has enabled the differentiation between peripheral monocytes and resident microglia, which was previously considered difficult. Thus, the evidence suggests that peripheral monocytes may contribute to the pathogenesis of seizures.     

Measurement of self-diffusion coefficients in Li ionic conductors by using the short-lived radiotracer of 8Li

For the effective use of short-lived radioactive beams, soon to be available at the Tokai Radioactive Ion Accelerator Complex, the authors have developed a radiotracer method for diffusion studies in solids. The experimental test was performed by the measurement of the diffusion coefficients of Li in a sample of the compound βLiAl using an α-emitting radiotracer of ⁸Li (T_1/2=0.84 s). It was found that the time-dependent yields of the α particles from the diffusing ⁸Li that was initially implanted in the sample could be used as a measure of the diffusivity of the tracer in a nondestructive way. The method was applied to measure the self-diffusion coefficients of Li in βLiGa, and for investigating how the Li diffusion in the Li ionic conductors is affected by the concentration of atomic defects (i.e., the existence of the atomic vacancies of Li and the defects in Ga sites that are replaced by Li).     

Dynamics of learning near singularities in layered networks

We explicitly analyze the trajectories of learning near singularities in hierarchical networks, such as multilayer perceptrons and radial basis function networks, which include permutation symmetry of hidden nodes, and show their general properties. Such symmetry induces singularities in their parameter space, where the Fisher information matrix degenerates and odd learning behaviors, especially the existence of plateaus in gradient descent learning, arise due to the geometric structure of singularity. We plot dynamic vector fields to demonstrate the universal trajectories of learning near singularities. The singularity induces two types of plateaus, the on-singularity plateau and the near-singularity plateau, depending on the stability of the singularity and the initial parameters of learning. The results presented in this letter are universally applicable to a wide class of hierarchical models. Detailed stability analysis of the dynamics of learning in radial basis function networks and multilayer perceptrons will be presented in separate work.