Large-Scale Elastic Net Regularized Linear Classification SVMs and Logistic Regression

Elastic Net Regularizers have shown much promise in designing sparse classifiers for linear classification. In this work, we propose an alternating optimization approach to solve the dual problems of elastic net regularized linear classification Support Vector Machines (SVMs) and logistic regression (LR). One of the sub-problems turns out to be a simple projection. The other sub-problem can be solved using dual coordinate descent methods developed for non-sparse L2-regularized linear SVMs and LR, without altering their iteration complexity and convergence properties. Experiments on very large datasets indicate that the proposed dual coordinate descent - projection (DCD-P) methods are fast and achieve comparable generalization performance after the first pass through the data, with extremely sparse models.

REDUCING COMPUTATIONAL EFFORT IN SOLVING GEOMECHANICS PROBLEMS WITH UPPER BOUND FINITE ELEMENTS LIMIT ANALYSIS AND LINEAR OPTIMIZATION

This paper presents a simple technique for reducing the computational effort while solving any geotechnical stability problem by using the upper bound finite element limit analysis and linear optimization. In the proposed method, the problem domain is discretized into a number of different regions in which a particular order (number of sides) of the polygon is chosen to linearize the Mohr-Coulomb yield criterion. A greater order of the polygon needs to be selected only in that region wherein the rate of the plastic strains becomes higher. The computational effort required to solve the problem with this implementation reduces considerably. By using the proposed method, the bearing capacity has been computed for smooth and rough strip footings and the results are found to be quite satisfactory.

Space-Vector-Based Hybrid Pulsewidth Modulation Techniques for a Three-Level Inverter

Novel switching sequences have been proposed recently for a neutral-point-clamped three-level inverter, controlled effectively as an equivalent two-level inverter. It is shown that the four novel sequences can be grouped into two pairs of sequences. Using each pair of sequences, a hybrid pulsewidth modulation (PWM) technique is proposed, which deploys the two sequences in appropriate spatial regions to reduce the current ripple. Further, a third hybrid PWM technique is proposed which uses all the five sequences (including the conventional sequence) in appropriate spatial regions. Each proposed hybrid PWM is shown, both analytically and experimentally, to outperform its constituent PWM methods in terms of harmonic distortion. In particular, the third proposed hybrid PWM reduces the total harmonic distortion considerably at low- and high-speed ranges of a constant volts-per-hertz induction motor drive, compared to centered space vector PWM.

Solving Axisymmetric Stability Problems by Using Upper Bound Finite Elements, Limit Analysis, and Linear Optimization

A numerical formulation has been proposed for solving an axisymmetric stability problem in geomechanics with upper bound limit analysis, finite elements, and linear optimization. The Drucker-Prager yield criterion is linearized by simulating a sphere with a circumscribed truncated icosahedron. The analysis considers only the velocities and plastic multiplier rates, not the stresses, as the basic unknowns. The formulation is simple to implement, and it has been employed for finding the collapse loads of a circular footing placed over the surface of a cohesive-frictional material. The formulation can be used to solve any general axisymmetric geomechanics stability problem.

A Nearly Constant Switching Frequency Current Error Space Vector Based Hysteresis Controller for an IM Drive with 12-sided Polygonal Voltage Space Vectors

In this paper, a current error space vector (CESV) based hysteresis controller for a 12-sided polygonal voltage space vector inverter fed induction motor (IM) drive is proposed, for the first time. An open-end winding configuration is used for the induction motor. The proposed controller uses parabolic boundary with generalized vector selection logic for all sectors. The drive scheme is first studied with a space vector based PWM (SVPWM) control and from this the current error space phasor boundary is obtained. This current error space phasor boundary is approximated with four parabolas and then the system is run with space phasor based hysteresis PWM controller by limiting the CESV within the parabolic boundary. The proposed controller has increased modulation range, absence of 5th and 7th order harmonics for the entire modulation range, nearly constant switching frequency, fast dynamic response with smooth transition to the over modulation region and a simple controller implementation.

Multi-site downscaling of maximum and minimum daily temperature using support vector machine

Climate change impact assessment studies involve downscaling large-scale atmospheric predictor variables (LSAPVs) simulated by general circulation models (GCMs) to site-scale meteorological variables. This article presents a least-square support vector machine (LS-SVM)-based methodology for multi-site downscaling of maximum and minimum daily temperature series. The methodology involves (1) delineation of sites in the study area into clusters based on correlation structure of predictands, (2) downscaling LSAPVs to monthly time series of predictands at a representative site identified in each of the clusters, (3) translation of the downscaled information in each cluster from the representative site to that at other sites using LS-SVM inter-site regression relationships, and (4) disaggregation of the information at each site from monthly to daily time scale using k-nearest neighbour disaggregation methodology. Effectiveness of the methodology is demonstrated by application to data pertaining to four sites in the catchment of Beas river basin, India. Simulations of Canadian coupled global climate model (CGCM3.1/T63) for four IPCC SRES scenarios namely A1B, A2, B1 and COMMIT were downscaled to future projections of the predictands in the study area. Comparison of results with those based on recently proposed multivariate multiple linear regression (MMLR) based downscaling method and multi-site multivariate statistical downscaling (MMSD) method indicate that the proposed method is promising and it can be considered as a feasible choice in statistical downscaling studies. The performance of the method in downscaling daily minimum temperature was found to be better when compared with that in downscaling daily maximum temperature. Results indicate an increase in annual average maximum and minimum temperatures at all the sites for A1B, A2 and B1 scenarios. The projected increment is high for A2 scenario, and it is followed by that for A1B, B1 and COMMIT scenarios. Projections, in general, indicated an increase in mean monthly maximum and minimum temperatures during January to February and October to December.

Quantum simulation using fidelity-profile optimization

Experimental quantum simulation of a Hamiltonian H requires unitary operator decomposition (UOD) of its evolution unitary U = exp(-iHt) in terms of native unitary operators of the experimental system. Here, using a genetic algorithm, we numerically evaluate the most generic UOD (valid over a continuous range of Hamiltonian parameters) of the unitary operator U, termed fidelity-profile optimization. The optimization is obtained by systematically evaluating the functional dependence of experimental unitary operators (such as single-qubit rotations and time-evolution unitaries of the system interactions) to the Hamiltonian (H) parameters. Using this technique, we have solved the experimental unitary decomposition of a controlled-phase gate (for any phase value), the evolution unitary of the Heisenberg XY interaction, and simulation of the Dzyaloshinskii-Moriya (DM) interaction in the presence of the Heisenberg XY interaction. Using these decompositions, we studied the entanglement dynamics of a Bell state in the DM interaction and experimentally verified the entanglement preservation procedure of Hou et al. Ann. Phys. (N.Y.) 327, 292 (2012)] in a nuclear magnetic resonance quantum information processor.

Freeform Skeletal Shape Optimization of Compliant Mechanisms

Compliant mechanisms are elastic continua used to transmit or transform force and motion mechanically. The topology optimization methods developed for compliant mechanisms also give the shape for a chosen parameterization of the design domain with a fixed mesh. However, in these methods, the shapes of the flexible segments in the resulting optimal solutions are restricted either by the type or the resolution of the design parameterization. This limitation is overcome in this paper by focusing on optimizing the skeletal shape of the compliant segments in a given topology. It is accomplished by identifying such segments in the topology and representing them using Bezier curves. The vertices of the Bezier control polygon are used to parameterize the shape-design space. Uniform parameter steps of the Bezier curves naturally enable adaptive finite element discretization of the segments as their shapes change. Practical constraints such as avoiding intersections with other segments, self-intersections, and restrictions on the available space and material, are incorporated into the formulation. A multi-criteria function from our prior work is used as the objective. Analytical sensitivity analysis for the objective and constraints is presented and is used in the numerical optimization. Examples are included to illustrate the shape optimization method.

Optimization of degree of sphericity of primary phase during cooling slope casting of A356 Al alloy: Taguchi method and regression analysis

The present work presents the results of experimental investigation of semi-solid rheocasting of A356 Al alloy using a cooling slope. The experiments have been carried out following Taguchi method of parameter design (orthogonal array of L-9 experiments). Four key process variables (slope angle, pouring temperature, wall temperature, and length of travel of the melt) at three different levels have been considered for the present experimentation. Regression analysis and analysis of variance (ANOVA) has also been performed to develop a mathematical model for degree of sphericity evolution of primary alpha-Al phase and to find the significance and percentage contribution of each process variable towards the final outcome of degree of sphericity, respectively. The best processing condition has been identified for optimum degree of sphericity (0.83) as A(3), B-3, C-2, D-1 i.e., slope angle of 60 degrees, pouring temperature of 650 degrees C, wall temperature 60 degrees C, and 500 mm length of travel of the melt, based on mean response and signal to noise ratio (SNR). ANOVA results shows that the length of travel has maximum impact on degree of sphericity evolution. The predicted sphericity obtained from the developed regression model and the values obtained experimentally are found to be in good agreement with each other. The sphericity values obtained from confirmation experiment, performed at 95% confidence level, ensures that the optimum result is correct and also the confirmation experiment values are within permissible limits. (c) 2014 Elsevier Ltd. All rights reserved.

Smoothed Functional Algorithms for Stochastic Optimization Using q-Gaussian Distributions

Smoothed functional (SF) schemes for gradient estimation are known to be efficient in stochastic optimization algorithms, especially when the objective is to improve the performance of a stochastic system However, the performance of these methods depends on several parameters, such as the choice of a suitable smoothing kernel. Different kernels have been studied in the literature, which include Gaussian, Cauchy, and uniform distributions, among others. This article studies a new class of kernels based on the q-Gaussian distribution, which has gained popularity in statistical physics over the last decade. Though the importance of this family of distributions is attributed to its ability to generalize the Gaussian distribution, we observe that this class encompasses almost all existing smoothing kernels. This motivates us to study SF schemes for gradient estimation using the q-Gaussian distribution. Using the derived gradient estimates, we propose two-timescale algorithms for optimization of a stochastic objective function in a constrained setting with a projected gradient search approach. We prove the convergence of our algorithms to the set of stationary points of an associated ODE. We also demonstrate their performance numerically through simulations on a queuing model.

Survey of vector-like fermion extensions of the Standard Model and their phenomenological implications

With the renewed interest in vector-like fermion extensions of the Standard Model, we present here a study of multiple vector-like theories and their phenomenological implications. Our focus is mostly on minimal flavor conserving theories that couple the vector-like fermions to the SM gauge fields and mix only weakly with SM fermions so as to avoid flavor problems. We present calculations for precision electroweak and vector-like state decays, which are needed to investigate compatibility with currently known data. We investigate the impact of vector-like fermions on Higgs boson production and decay, including loop contributions, in a wide variety of vector-like extensions and their parameter spaces.

A perturbed martingale approach to global optimization

A new global stochastic search, guided mainly through derivative-free directional information computable from the sample statistical moments of the design variables within a Monte Carlo setup, is proposed. The search is aided by imparting to the directional update term additional layers of random perturbations referred to as `coalescence' and `scrambling'. A selection step, constituting yet another avenue for random perturbation, completes the global search. The direction-driven nature of the search is manifest in the local extremization and coalescence components, which are posed as martingale problems that yield gain-like update terms upon discretization. As anticipated and numerically demonstrated, to a limited extent, against the problem of parameter recovery given the chaotic response histories of a couple of nonlinear oscillators, the proposed method appears to offer a more rational, more accurate and faster alternative to most available evolutionary schemes, prominently the particle swarm optimization. (C) 2014 Elsevier B.V. All rights reserved.

Space vector based modulation scheme for reducing capacitor RMS current in three-level diode-clamped inverter

The DC capacitor is an important component in a voltage source inverter.The RMS current flowing through the capacitor determines the capacitor size and losses. The losses, in turn, influence the capacitor life. This paper proposes a space vector based modulation strategy for reducing the capacitor RMS current in a three-level diode-clamped inverter. An analytical closed-form expression is derived for the DC capacitor RMS current with the proposed PWM strategy. The analytical expression is validated through simulations and also experimentally. Theoretical and experimental results are presented, comparing the proposed strategy with conventional space vector PWM (CSVPWM). It is shown that the proposed strategy reduces the capacitor RMS current significantly at high modulation indices and high power factors. (C) 2014 Elsevier B.V. All rights reserved.

Newton-based stochastic optimization using q-Gaussian smoothed functional algorithms

We present the first q-Gaussian smoothed functional (SF) estimator of the Hessian and the first Newton-based stochastic optimization algorithm that estimates both the Hessian and the gradient of the objective function using q-Gaussian perturbations. Our algorithm requires only two system simulations (regardless of the parameter dimension) and estimates both the gradient and the Hessian at each update epoch using these. We also present a proof of convergence of the proposed algorithm. In a related recent work (Ghoshdastidar, Dukkipati, & Bhatnagar, 2014), we presented gradient SF algorithms based on the q-Gaussian perturbations. Our work extends prior work on SF algorithms by generalizing the class of perturbation distributions as most distributions reported in the literature for which SF algorithms are known to work turn out to be special cases of the q-Gaussian distribution. Besides studying the convergence properties of our algorithm analytically, we also show the results of numerical simulations on a model of a queuing network, that illustrate the significance of the proposed method. In particular, we observe that our algorithm performs better in most cases, over a wide range of q-values, in comparison to Newton SF algorithms with the Gaussian and Cauchy perturbations, as well as the gradient q-Gaussian SF algorithms. (C) 2014 Elsevier Ltd. All rights reserved.

An Eigen Approach to Transmit Beamforming in Wireless Networks With Single Antenna Receivers

In this paper, we propose an eigen framework for transmit beamforming for single-hop and dual-hop network models with single antenna receivers. In cases where number of receivers is not more than three, the proposed Eigen approach is vastly superior in terms of ease of implementation and computational complexity compared with the existing convex-relaxation-based approaches. The essential premise is that the precoding problems can be posed as equivalent optimization problems of searching for an optimal vector in the joint numerical range of Hermitian matrices. We show that the latter problem has two convex approximations: the first one is a semi-definite program that yields a lower bound on the solution, and the second one is a linear matrix inequality that yields an upper bound on the solution. We study the performance of the proposed and existing techniques using numerical simulations.