A fast quasi-Newton method for semi-supervised SVM

Due to its wide applicability, semi-supervised learning is an attractive method for using unlabeled data in classification. In this work, we present a semi-supervised support vector classifier that is designed using quasi-Newton method for nonsmooth convex functions. The proposed algorithm is suitable in dealing with very large number of examples and features. Numerical experiments on various benchmark datasets showed that the proposed algorithm is fast and gives improved generalization performance over the existing methods. Further, a non-linear semi-supervised SVM has been proposed based on a multiple label switching scheme. This non-linear semi-supervised SVM is found to converge faster and it is found to improve generalization performance on several benchmark datasets. (C) 2010 Elsevier Ltd. All rights reserved.

Ultrasonic evaluation of delamination in quasi-isotropic CFRP laminates subjected to low-velocity impact

Ultrasonic C-Scan is used very often to detect flaws and defects in the composite components resulted during fabrication and damages resulting from service conditions. Evaluation and characterization of defects and damages of composites require experience and good understanding of the material as they are distinctly different in composition and behavior as compared to conventional metallic materials. The failure mechanisms in composite materials are quite complex. They involve the interaction of matrix cracking, fiber matrix interface debonding, fiber pullout, fiber fracture and delamination. Generally all of them occur making the stress and failure analysis very complex. Under low-velocity impact loading delamination is observed to be a major failure mode. In composite materials the ultrasonic waves suffer high acoustic attenuation and scattering effect, thus making data interpretation difficult. However these difficulties can be overcome to a greater extent by proper selection of probe, probe parameter settings like pulse width, pulse amplitude, pulse repetition rate, delay, blanking, gain etc., and data processing which includes image processing done on the image obtained by the C-Scan.

Effect of scaling on the non-quasi-static behaviour of the MOSFET for RF ICs

The effect of scaling (1 μm to 0.09 μm) on the non-quasi-static (NQS) behaviour of the MOSFET has been studied using process and device simulation. It is shown that under fixed gate (Vgs) and drain (Vds) bias voltages, the NQS transition frequency (fNQS) scales as 1/Leff rather than 1/L2eff due to the velocity saturation effect. However, under the practical scaling guidelines, considering the scaling of supply voltage as well, fNQS shows a turn around effect at the sub 100 nm regime. The relation between unity gain frequency (ft) and fNQS is also evaluated and it is shown that ft and fNQS have similar trends with scaling.

Quasi-static crack tip fields in rate-sensitive FCC single crystals

In this work, the effects of loading rate, material rate sensitivity and constraint level on quasi-static crack tip fields in a FCC single crystal are studied. Finite element simulations are performed within a mode I, plane strain modified boundary layer framework by prescribing the two term (K-T) elastic crack tip field as remote boundary conditions. The material is assumed to obey a rate-dependent crystal plasticity theory. The orientation of the single crystal is chosen so that the crack surface coincides with the crystallographic (010) plane and the crack front lies along 101] direction. Solutions corresponding to different stress intensity rates K., T-stress values and strain rate exponents m are obtained. The results show that the stress levels ahead of the crack tip increase with K. which is accompanied by gradual shrinking of the plastic zone size. However, the nature of the shear band patterns around the crack tip is not affected by the loading rate. Further, it is found that while positive T-stress enhances the opening and hydrostatic stress levels ahead of crack tip, they are considerably reduced with imposition of negative T-stress. Also, negative T-stress promotes formation of shear bands in the forward sector ahead of the crack tip and suppresses them behind the tip.

Smoothed functional and Quasi-Newton algorithms for routing in multi-stage queueing network with constraints

We consider the problem of optimal routing in a multi-stage network of queues with constraints on queue lengths. We develop three algorithms for probabilistic routing for this problem using only the total end-to-end delays. These algorithms use the smoothed functional (SF) approach to optimize the routing probabilities. In our model all the queues are assumed to have constraints on the average queue length. We also propose a novel quasi-Newton based SF algorithm. Policies like Join Shortest Queue or Least Work Left work only for unconstrained routing. Besides assuming knowledge of the queue length at all the queues. If the only information available is the expected end-to-end delay as with our case such policies cannot be used. We also give simulation results showing the performance of the SF algorithms for this problem.

Quasi-based hierarchical clustering for land cover mapping using satellite images

This paper presents an improved hierarchical clustering algorithm for land cover mapping problem using quasi-random distribution. Initially, Niche Particle Swarm Optimization (NPSO) with pseudo/quasi-random distribution is used for splitting the data into number of cluster centers by satisfying Bayesian Information Criteria (BIC). Themain objective is to search and locate the best possible number of cluster and its centers. NPSO which highly depends on the initial distribution of particles in search space is not been exploited to its full potential. In this study, we have compared more uniformly distributed quasi-random with pseudo-random distribution with NPSO for splitting data set. Here to generate quasi-random distribution, Faure method has been used. Performance of previously proposed methods namely K-means, Mean Shift Clustering (MSC) and NPSO with pseudo-random is compared with the proposed approach - NPSO with quasi distribution(Faure). These algorithms are used on synthetic data set and multi-spectral satellite image (Landsat 7 thematic mapper). From the result obtained we conclude that use of quasi-random sequence with NPSO for hierarchical clustering algorithm results in a more accurate data classification.

Electronic band structure and Fermi surfaces of the quasi-two-dimensional monophosphate tungsten bronze, P4W12O44

The electronic structure of quasi-two-dimensional monophosphate tungsten bronze, P4W12O44, has been investigated by high-resolution angle-resolved photoemission spectroscopy and density functional theoretical calculations. Experimental electron-like bands around Gamma point and Fermi surfaces have similar shapes as predicted by calculations. Fermi surface mapping at different temperatures shows a depletion of density of states at low temperature in certain flat portions of the Fermi surfaces. These flat portions of the Fermi surfaces satisfy the partial nesting condition with incommensurate nesting vectors q(1) and q(2), which leads to the formation of charge density waves in this phosphate tungsten bronzes. The setting up of charge density wave in these bronzes can well explain the anomaly observed in its transport properties. Copyright (C) EPLA, 2014

Raman Evidence for Coupling of Superconducting Quasi-Particles with a Phonon and Crystal Field Excitation in Superconductor Ce0.6Y0.4FeAsO0.8F0.2

We report inelastic light scattering experiments on superconductor Ce0.6Y0.4FeAsO0.8F0.2 from 4K to 300K covering the superconducting transition temperature T-c similar to 48.6K. A strong evidence of the superconductivity induced phonon renormalization for the A(1g) phonon mode near 150cm(-1) associated with the Ce/Y vibrations is observed as reflected in the anomalous red-shift and decrease in the linewidth below T-c. Invoking the coupling of this mode with the superconducting gap, the superconducting gap (2 Delta) at zero temperature is estimated to be similar to 20meV i.e the ratio 2 Delta(0)/k(B)T(c) is similar to 5, suggesting Ce0.6Y0.4FeAsO0.8F0.2 to belong to the class of strong coupling superconductors. In addition, the mode near 430cm(-1) associated with Ce3+ crystal field excitation also shows anomalous increase in its linewidth below T-c suggesting strong coupling between crystal field excitation and the superconducting quasi-particles.

Synthesis and photocatalytic performance of quasi-fibrous ZnO

The combustion of oxidizer zinc nitrate and fuel oxalic acid results in quasi-fibrous zinc oxide. The processing parameters including oxidizer to fuel ratio, time and temperature were optimized for the resultant crystal structure and morphology. Pure hexagonal phase formation does not depend on the fuel ratio, but a stoichiometric ratio of oxidizer to fuel at 450 degrees C and 30 min results in highly crystalline ZnO with 3 mu m length and 0.5 mu m width. This quasi-fiber originates from partial fusion of near spherical, similar to 60 nm particles during the rapid rate of reaction in the combustion process. Transmission electron microscopic analysis confirms the anisotropic primary particle orientation and pore distribution within the developed quasi-fibrous particles. The degradation of methyl orange was assessed by degrading the dye in the presence of the synthesized ZnO (2.95 eV) under both UV and visible light. Quasi-fibrous zinc oxide exhibits effective photocatalytic efficiency under visible light irradiation.

Bearing capacity of circular footings over rock mass by using axisymmetric quasi lower bound finite element limit analysis

The ultimate bearing capacity of a circular footing, placed over rock mass, is evaluated by using the lower bound theorem of the limit analysis in conjunction with finite elements and nonlinear optimization. The generalized Hoek-Brown (HB) failure criterion, but by keeping a constant value of the exponent, alpha = 0.5, was used. The failure criterion was smoothened both in the meridian and pi planes. The nonlinear optimization was carried out by employing an interior point method based on the logarithmic barrier function. The results for the obtained bearing capacity were presented in a non-dimensional form for different values of GSI, m(i), sigma(ci)/(gamma b) and q/sigma(ci). Failure patterns were also examined for a few cases. For validating the results, computations were also performed for a strip footing as well. The results obtained from the analysis compare well with the data reported in literature. Since the equilibrium conditions are precisely satisfied only at the centroids of the elements, not everywhere in the domain, the obtained lower bound solution will be approximate not true. (C) 2015 Elsevier Ltd. All rights reserved.

Quasi-geostrophic dynamics in the presence of moisture gradients

The derivation of a quasi-geostrophic system from the rotating shallow-water equations on a midlatitude -plane coupled with moisture is presented. Condensation is prescribed to occur whenever the moisture at a point exceeds a prescribed saturation value. It is seen that a slow condensation time-scale is required to obtain a consistent set of equations at leading order. Further, since the advecting wind fields are geostrophic, changes in moisture (and hence precipitation) occur only via non-divergent mechanisms. Following observations, a saturation profile with gradients in the zonal and meridional directions is prescribed. A purely meridional gradient has the effect of slowing down the dry Rossby waves, through a reduction in the equivalent gradient' of the background potential vorticity. A large-scale unstable moist mode results on the inclusion of a zonal gradient by itself, or in conjunction with a meridional moisture gradient. For gradients that are are representative of the atmosphere, the most unstable moist mode propagates zonally in the direction of increasing moisture, matures over an intraseasonal time-scale and has small phase speed.

Reply to Discussion on ``Bearing capacity of circular footings over rock mass by using axisymmetric quasi lower bound finite element limit analysis'' Comput. Geotech. 70 (2015) 138-149]

A discussion has been provided for the comments raised by the discusser (Clausen, 2015)1] on the article recently published by the authors (Chakraborty and Kumar, 2015). The effect of exponent alpha for values of GSI approximately smaller than 30 becomes more critical. On the other hand, for greater values of GSI, the results obtained by the authors earlier remain primarily independent of alpha and can be easily used. (C) 2015 Elsevier Ltd. All rights reserved.

On the Geometrically Dependent Quasi-Saturation and g(m) Reduction in Advanced DeMOS Transistors

This paper reveals an early quasi-saturation (QS) effect attributed to the geometrical parameters in shallow trench isolation-type drain-extended MOS (STI-DeMOS) transistors in advanced CMOS technologies. The quasi-saturation effect leads to serious g(m) reduction in STI-DeMOS. This paper investigates the nonlinear resistive behavior of the drain-extended region and its impact on the particular behavior of the STI-DeMOS transistor. In difference to vertical DMOS or lateral DMOS structures, STI-DeMOS exhibits three distinct regions of the drain extension. A complete understanding of the physics in these regions and their impact on the QS behavior are developed in this paper. An optimization strategy is shown for an improved g(m) device in a state-of-the-art 28-nm CMOS technology node.