A Hybrid Forward Algorithm for RBF Neural Network Construction

This paper proposes a novel hybrid forward algorithm (HFA) for the construction of radial basis function (RBF) neural networks with tunable nodes. The main objective is to efficiently and effectively produce a parsimonious RBF neural network that generalizes well. In this study, it is achieved through simultaneous network structure determination and parameter optimization on the continuous parameter space. This is a mixed integer hard problem and the proposed HFA tackles this problem using an integrated analytic framework, leading to significantly improved network performance and reduced memory usage for the network construction. The computational complexity analysis confirms the efficiency of the proposed algorithm, and the simulation results demonstrate its effectiveness

A fast hybrid algorithm for exoplanetary transit searches

We present a fast and efficient hybrid algorithm for selecting exoplanetary candidates from wide-field transit surveys. Our method is based on the widely used SysRem and Box Least-Squares (BLS) algorithms. Patterns of systematic error that are common to all stars on the frame are mapped and eliminated using the SysRem algorithm. The remaining systematic errors caused by spatially localized flat-fielding and other errors are quantified using a boxcar-smoothing method. We show that the dimensions of the search-parameter space can be reduced greatly by carrying out an initial BLS search on a coarse grid of reduced dimensions, followed by Newton-Raphson refinement of the transit parameters in the vicinity of the most significant solutions. We illustrate the method's operation by applying it to data from one field of the SuperWASP survey, comprising 2300 observations of 7840 stars brighter than V = 13.0. We identify 11 likely transit candidates. We reject stars that exhibit significant ellipsoidal variations caused indicative of a stellar-mass companion. We use colours and proper motions from the Two Micron All Sky Survey and USNO-B1.0 surveys to estimate the stellar parameters and the companion radius. We find that two stars showing unambiguous transit signals pass all these tests, and so qualify for detailed high-resolution spectroscopic follow-up.

Cutaneous Grooves: Composing for the Sense of Touch

What if the traditional relationship between touch and music was essentially turned upside down, making the tactile sensation the aesthetic end? This paper presents a novel coupling of haptics technology and music, introducing the notion of tactile composition or aesthetic composition for the sense of touch. A system that facilitates the composition and perception of intricate, musically structured spatio-temporal patterns of vibration on the surface of the body is described. Relevant work from disciplines including sensory substitution, electronic musical instrument design, simulation design, entertainment technology, and visual music is considered. The psychophysical parameter space for our sense of touch is summarized and the building blocks of a compositional language for touch are explored. A series of concerts held for the skin and ears is described, as well as some of the lessons learned along the way. In conclusion, some potential evolutionary branches of tactile composition are posited.

Distribution-based adaptive bounding genetic algorithm for continuous optimisation problems

A new search-space-updating technique for genetic algorithms is proposed for continuous optimisation problems. Other than gradually reducing the search space during the evolution process with a fixed reduction rate set ‘a priori’, the upper and the lower boundaries for each variable in the objective function are dynamically adjusted based on its distribution statistics. To test the effectiveness, the technique is applied to a number of benchmark optimisation problems in comparison with three other techniques, namely the genetic algorithms with parameter space size adjustment (GAPSSA) technique [A.B. Djurišic, Elite genetic algorithms with adaptive mutations for solving continuous optimization problems – application to modeling of the optical constants of solids, Optics Communications 151 (1998) 147–159], successive zooming genetic algorithm (SZGA) [Y. Kwon, S. Kwon, S. Jin, J. Kim, Convergence enhanced genetic algorithm with successive zooming method for solving continuous optimization problems, Computers and Structures 81 (2003) 1715–1725] and a simple GA. The tests show that for well-posed problems, existing search space updating techniques perform well in terms of convergence speed and solution precision however, for some ill-posed problems these techniques are statistically inferior to a simple GA. All the tests show that the proposed new search space update technique is statistically superior to its counterparts.

Learning soft computing control strategies in a modular neural network architecture

Modelling and control of nonlinear dynamical systems is a challenging problem since the dynamics of such systems change over their parameter space. Conventional methodologies for designing nonlinear control laws, such as gain scheduling, are effective because the designer partitions the overall complex control into a number of simpler sub-tasks. This paper describes a new genetic algorithm based method for the design of a modular neural network (MNN) control architecture that learns such partitions of an overall complex control task. Here a chromosome represents both the structure and parameters of an individual neural network in the MNN controller and a hierarchical fuzzy approach is used to select the chromosomes required to accomplish a given control task. This new strategy is applied to the end-point tracking of a single-link flexible manipulator modelled from experimental data. Results show that the MNN controller is simple to design and produces superior performance compared to a single neural network (SNN) controller which is theoretically capable of achieving the desired trajectory. (C) 2003 Elsevier Ltd. All rights reserved.

Line-profile tomography of exoplanet transits - II. A gas-giant planet transiting a rapidly rotating A5 star

Most of our knowledge of extrasolar planets rests on precise radial-velocity measurements, either for direct detection or for confirmation of the planetary origin of photometric transit signals. This has limited our exploration of the parameter space of exoplanet hosts to solar- and later-type, sharp-lined stars. Here we extend the realm of stars with known planetary companions to include hot, fast-rotating stars. Planet-like transits have previously been reported in the light curve obtained by the SuperWASP survey of the A5 star HD15082 (WASP-33 V = 8.3, v sini = 86 km s-1). Here we report further photometry and time-series spectroscopy through three separate transits, which we use to confirm the existence of a gas-giant planet with an orbital period of 1.22d in orbit around HD15082. From the photometry and the properties of the planet signal travelling through the spectral line profiles during the transit, we directly derive the size of the planet, the inclination and obliquity of its orbital plane and its retrograde orbital motion relative to the spin of the star. This kind of analysis opens the way to studying the formation of planets around a whole new class of young, early-type stars, hence under different physical conditions and generally in an earlier stage of formation than in sharp-lined late-type stars. The reflex orbital motion of the star caused by the transiting planet is small, yielding an upper mass limit of 4.1MJupiter on the planet. We also find evidence of a third body of substellar mass in the system, which may explain the unusual orbit of the transiting planet. In HD 15082, the stellar line profiles also show evidence of non-radial pulsations, clearly distinct from the planetary transit signal. This raises the intriguing possibility that tides raised by the close-in planet may excite or amplify the pulsations in such stars.

Two-stage mixed discrete-continuous identification of radial basis function (RBF) neural models for nonlinear systems

The identification of nonlinear dynamic systems using radial basis function (RBF) neural models is studied in this paper. Given a model selection criterion, the main objective is to effectively and efficiently build a parsimonious compact neural model that generalizes well over unseen data. This is achieved by simultaneous model structure selection and optimization of the parameters over the continuous parameter space. It is a mixed-integer hard problem, and a unified analytic framework is proposed to enable an effective and efficient two-stage mixed discrete-continuous; identification procedure. This novel framework combines the advantages of an iterative discrete two-stage subset selection technique for model structure determination and the calculus-based continuous optimization of the model parameters. Computational complexity analysis and simulation studies confirm the efficacy of the proposed algorithm.

Discretization of highly persistent correlated AR(1) shocks

The finite state Markov-chain approximation methods developed by Tauchen (1986) and Tauchen and Hussey (1991) are widely used in economics, finance and econometrics to solve functional equations in which state variables follow autoregressive processes. For highly persistent processes, the methods require a large number of discrete values for the state variables to produce close approximations which leads to an undesirable reduction in computational speed, especially in a multivariate case. This paper proposes an alternative method of discretizing multivariate autoregressive processes. This method can be treated as an extension of Rouwenhorst's (1995) method which, according to our finding, outperforms the existing methods in the scalar case for highly persistent processes. The new method works well as an approximation that is much more robust to the number of discrete values for a wide range of the parameter space.

Acoustic solitary waves in dusty and/or multi-ion plasmas with cold, adiabatic, and hot constituents

Large nonlinear acoustic waves are discussed in a four-component plasma, made up of two superhot isothermal species, and two species with lower thermal velocities, being, respectively, adiabatic and cold. First a model is considered in which the isothermal species are electrons and ions, while the cooler species are positive and/or negative dust. Using a Sagdeev pseudopotential formalism, large dust-acoustic structures have been studied in a systematic way, to delimit the compositional parameter space in which they can be found, without restrictions on the charges and masses of the dust species and their charge signs. Solitary waves can only occur for nonlinear structure velocities smaller than the adiabatic dust thermal velocity, leading to a novel dust-acoustic-like mode based on the interplay between the two dust species. If the cold and adiabatic dust are oppositely charged, only solitary waves exist, having the polarity of the cold dust, their parameter range being limited by infinite compression of the cold dust. However, when the charges of the cold and adiabatic species have the same sign, solitary structures are limited for increasing Mach numbers successively by infinite cold dust compression, by encountering the adiabatic dust sonic point, and by the occurrence of double layers. The latter have, for smaller Mach numbers, the same polarity as the charged dust, but switch at the high Mach number end to the opposite polarity. Typical Sagdeev pseudopotentials and solitary wave profiles have been presented. Finally, the analysis has nowhere used the assumption that the dust would be much more massive than the ions and hence, one or both dust species can easily be replaced by positive and/or negative ions and the conclusions will apply to that plasma model equally well. This would cover a number of different scenarios, such as, for example, very hot electrons and ions, together with a mix of adiabatic ions and dust (of either polarity) or a very hot electron-positron mix, together with a two-ion mix or together with adiabatic ions and cold dust (both of either charge sign), to name but some of the possible plasma compositions.

Dust ion acoustic solitons in a plasma with kappa-distributed electrons

Dust ion acoustic solitons in an unmagnetized dusty plasma comprising cold dust particles, adiabatic fluid ions, and electrons satisfying a kappa distribution are investigated using both small amplitude and arbitrary amplitude techniques. Their existence domain is discussed in the parameter space of Mach number M and electron density fraction f over a wide range of values of kappa. For all kappa > 3/2, including the Maxwellian distribution, negative dust supports solitons of both polarities over a range in f. In that region of parameter space solitary structures of finite amplitude can be obtained even at the lowest Mach number, the acoustic speed, for all kappa. These cannot be found from small amplitude theories. This surprising behavior is investigated, and it is shown that f(c), the value of f at which the KdV coefficient A vanishes, plays a critical role. In the presence of positive dust, only positive potential solitons are found. (C) 2010 American Institute of Physics. [doi: 10.1063/1.3400229]

Towards accurate estimation of the proportion of true null hypotheses in multiple testing

Background

Biomedical researchers are now often faced with situations where it is necessary to test a large number of hypotheses simultaneously, eg, in comparative gene expression studies using high-throughput microarray technology. To properly control false positive errors the FDR (false discovery rate) approach has become widely used in multiple testing. The accurate estimation of FDR requires the proportion of true null hypotheses being accurately estimated. To date many methods for estimating this quantity have been proposed. Typically when a new method is introduced, some simulations are carried out to show the improved accuracy of the new method. However, the simulations are often very limited to covering only a few points in the parameter space.

Results

Here I have carried out extensive in silico experiments to compare some commonly used methods for estimating the proportion of true null hypotheses. The coverage of these simulations is unprecedented thorough over the parameter space compared to typical simulation studies in the literature. Thus this work enables us to draw conclusions globally as to the performance of these different methods. It was found that a very simple method gives the most accurate estimation in a dominantly large area of the parameter space. Given its simplicity and its overall superior accuracy I recommend its use as the first choice for estimating the proportion of true null hypotheses in multiple testing.

Transonic Aeroelastic Stability Analysis Using a Kriging-Based Schur Complement Formulation

A method is described to allow searches for transonic aeroelastic instability of realistically sized aircraft models in multidimensional parameter spaces when computational fluid dynamics are used to model the aerodynamics. Aeroelastic instability is predicted from a small nonlinear eigenvalue problem. The approximation of the computationally expensive interaction term modeling the fluid response is formulated to allow the automated and blind search for aeroelastic instability. The approximation uses a kriging interpolation of exact numerical samples covering the parameter space. The approach, demonstrated for the Goland wing and the multidisciplinary optimization transport wing, results in stability analyses over whole flight envelopes at an equivalent cost of several steady-state simulations.

Memory-keeping effects and forgetfulness in the dynamics of a qubit coupled to a spin chain

Using recently proposed measures for non-Markovianity [H.-P. Breuer, E. M. Laine, and J. Piilo, Phys. Rev. Lett. 103, 210401 (2009)], we study the dynamics of a qubit coupled to a spin environment via an energy-exchange mechanism. We show the existence of a point, in the parameter space of the system, where the qubit dynamics is effectively Markovian and that such a point separates two regions with completely different dynamical behaviors. Indeed, our study demonstrates that the qubit evolution can in principle be tuned from a perfectly forgetful one to a deep non-Markovian regime where the qubit is strongly affected by the dynamical backaction of the environmental spins. By means of a theoretical quantum process tomography analysis, we provide a complete and intuitive characterization of the qubit channel.

Determining the parametric effectiveness of a CAD model

The motivation for this paper is to present an approach for rating the quality of the parameters in a computer-aided design model for use as optimization variables. Parametric Effectiveness is computed as the ratio of change in performance achieved by perturbing the parameters in the optimum way, to the change in performance that would be achieved by allowing the boundary of the model to move without the constraint on shape change enforced by the CAD parameterization. The approach is applied in this paper to optimization based on adjoint shape sensitivity analyses. The derivation of parametric effectiveness is presented for optimization both with and without the constraint of constant volume. In both cases, the movement of the boundary is normalized with respect to a small root mean squared movement of the boundary. The approach can be used to select an initial search direction in parameter space, or to select sets of model parameters which have the greatest ability to improve model performance. The approach is applied to a number of example 2D and 3D FEA and CFD problems.

Synthetic X-ray spectra for simulations of the dynamics of an accretion flow irradiated by a quasar

Ultraviolet and X-ray observations show evidence of outflowing gas around many active galactic nuclei. It has been proposed that some of these outflows are driven off gas infalling towards the central supermassive black hole. We perform radiative transfer calculations to compute the gas ionization state and the emergent X-ray spectra for both two- and three-dimensional (3D) hydrodynamical simulations of this outflow-from-inflow scenario. By comparison with observations, our results can be used to test the theoretical models and guide future numerical simulations. We predict both absorption and emission features, most of which are formed in a polar funnel of relatively dense (10 -10 g cm ) outflowing gas. This outflow causes strong absorption for observer orientation angles of ?35°. Particularly in 3D, the strength of this absorption varies significantly for different lines of sight owing to the fragmentary structure of the gas flow. Although infalling material occupies a large fraction of the simulation volume, we do not find that it imprints strong absorption features in the X-ray spectra since the ionization state is predicted to be very high. Thus, an absence of observed inflow absorption features does not exclude the models. The main spectroscopic consequence of the infalling gas is a Compton-scattered continuum component that partially re-fills the absorption features caused by the outflowing polar funnel. Fluorescence and scattering in the outflow are predicted to give rise to several emission features including a multicomponent Fe Ka emission complex for all observer orientations. For the hydrodynamical simulations considered, we predict both ionization states and column densities for the outflowing gas that are too high to be quantitatively consistent with well-observed X-ray absorption systems. Nevertheless, our results are qualitatively encouraging and further exploration of the model parameter space is warranted. Higher resolution hydrodynamic simulations are needed to determine whether the outflows fragment on scales unresolved in our current study, which may yield the denser lower ionization material that could reconcile the models and the observations. © 2012 The Authors Monthly Notices of the Royal Astronomical Society © 2012 RAS.