Improved conjugate gradient implementation for least squares support vector machines

As a promising method for pattern recognition and function estimation, least squares support vector machines (LS-SVM) express the training in terms of solving a linear system instead of a quadratic programming problem as for conventional support vector machines (SVM). In this paper, by using the information provided by the equality constraint, we transform the minimization problem with a single equality constraint in LS-SVM into an unconstrained minimization problem, then propose reduced formulations for LS-SVM. By introducing this transformation, the times of using conjugate gradient (CG) method, which is a greatly time-consuming step in obtaining the numerical solution, are reduced to one instead of two as proposed by Suykens et al. (1999). The comparison on computational speed of our method with the CG method proposed by Suykens et al. and the first order and second order SMO methods on several benchmark data sets shows a reduction of training time by up to 44%. (C) 2011 Elsevier B.V. All rights reserved.

Semi batch learning with store management using enhanced conjugate gradient

This paper explores the performance of sliding-window based training, termed as semi batch, using multilayer perceptron (MLP) neural network in the presence of correlated data. The sliding window training is a form of higher order instantaneous learning strategy without the need of covariance matrix, usually employed for modeling and tracking purposes. Sliding-window framework is implemented to combine the robustness of offline learning algorithms with the ability to track online the underlying process of a function. This paper adopted sliding window training with recent advances in conjugate gradient direction with application of data store management e.g. simple distance measure, angle evaluation and the novel prediction error test. The simulation results show the best convergence performance is gained by using store management techniques. © 2012 Springer-Verlag.

R-matrix-Floquet theory of multiphoton processes:VIII. A linear equations method

A new linear equations method for calculating the R-matrix, which arises in the R-matrix-Floquet theory of multiphoton processes, is introduced. This method replaces the diagonalization of the Floquet Hamiltonian matrix by the solution of a set of linear simultaneous equations which are solved, in the present work, by the conjugate gradient method. This approach uses considerably less computer memory and can be readily ported onto parallel computers. It will thus enable much larger problems of current interest to be treated. This new method is tested by applying it to three-photon ionization of helium at frequencies where double resonances with a bound state and autoionizing states are important. Finally, an alternative linear equations method, which avoids the explicit calculation of the R-matrix by incorporating the boundary conditions directly, is described in an appendix.

Evaluating Asymmetric Multicore Systems-on-Chip using Iso-Metrics

The end of Dennard scaling has pushed power consumption into a first order concern for current systems, on par with performance. As a result, near-threshold voltage computing (NTVC) has been proposed as a potential means to tackle the limited cooling capacity of CMOS technology. Hardware operating in NTV consumes significantly less power, at the cost of lower frequency, and thus reduced performance, as well as increased error rates. In this paper, we investigate if a low-power systems-on-chip, consisting of ARM's asymmetric big.LITTLE technology, can be an alternative to conventional high performance multicore processors in terms of power/energy in an unreliable scenario. For our study, we use the Conjugate Gradient solver, an algorithm representative of the computations performed by a large range of scientific and engineering codes.

Performance and Fault Tolerance of Preconditioned Iterative Solvers on Low-Power ARM Architectures

As the complexity of computing systems grows, reliability and energy are two crucial challenges asking for holistic solutions. In this paper, we investigate the interplay among concurrency, power dissipation, energy consumption and voltage-frequency scaling for a key numerical kernel for the solution of sparse linear systems. Concretely, we leverage a task-parallel implementation of the Conjugate Gradient method, equipped with an state-of-the-art pre-conditioner embedded in the ILUPACK software, and target a low-power multi core processor from ARM.In addition, we perform a theoretical analysis on the impact of a technique like Near Threshold Voltage Computing (NTVC) from the points of view of increased hardware concurrency and error rate.

Segmentation of optic disc in retinal images using an improved gradient vector flow algorithm

Image segmentation plays an important role in the analysis of retinal images as the extraction of the optic disk provides important cues for accurate diagnosis of various retinopathic diseases. In recent years, gradient vector flow (GVF) based algorithms have been used successfully to successfully segment a variety of medical imagery. However, due to the compromise of internal and external energy forces within the resulting partial differential equations, these methods can lead to less accurate segmentation results in certain cases. In this paper, we propose the use of a new mean shift-based GVF segmentation algorithm that drives the internal/external energies towards the correct direction. The proposed method incorporates a mean shift operation within the standard GVF cost function to arrive at a more accurate segmentation. Experimental results on a large dataset of retinal images demonstrate that the presented method optimally detects the border of the optic disc.

CH abundance gradient in TMC-1

Aims. The aim of this study is to examine if the well-known chemical gradient in TMC-1 is reflected in the amount of rudimentary forms of carbon available in the gas-phase. As a tracer we use the CH radical which is supposed to be well correlated with carbon atoms and simple hydrocarbon ions. Methods. We observed the 9-cm ?-doubling lines of CH along the dense filament of TMC-1. The CH column densities were compared with the total H2 column densities derived using the 2MASS NIR data and previously published SCUBA maps and with OH column densities derived using previous observations with Effelsberg. We also modelled the chemical evolution of TMC-1 adopting physical conditions typical of dark clouds using the UMIST Database for Astrochemistry gas-phase reaction network to aid the interpretation of the observed OH/CH abundance ratios. Results. The CH column density has a clear peak in the vicinity of the cyanopolyyne maximum of TMC-1. The fractional CH abundance relative to H2 increases steadily from the northwestern end of the filament where it lies around 1.0 × 10-8 , to the southeast where it reaches a value of 2.0 × 10-8. The OH and CH column densities are well correlated, and we obtained OH/CH abundance ratios of ~16–20. These values are clearly larger than what has been measured recently in diffuse interstellar gas and is likely to be related to C to CO conversion at higher densities. The good correlation between CH and OH can be explained by similar production and destruction pathways. We suggest that the observed CH and OH abundance gradients are mainly due to enhanced abundances in a low-density envelope which becomes more prominent in the southeastern part and seems to continue beyond the dense filament. Conclusions. An extensive envelope probably signifies an early stage of dynamical evolution, and conforms with the detection of a large CH abundance in the southeastern part of the cloud. The implied presence of other simple forms of carbon in the gas phase provides a natural explanation for the observation of “early-type” molecules in this region.

A New Gradient Descent Approach for Local Learning of Fuzzy Neural Models

The majority of reported learning methods for Takagi-Sugeno-Kang fuzzy neural models to date mainly focus on the improvement of their accuracy. However, one of the key design requirements in building an interpretable fuzzy model is that each obtained rule consequent must match well with the system local behaviour when all the rules are aggregated to produce the overall system output. This is one of the distinctive characteristics from black-box models such as neural networks. Therefore, how to find a desirable set of fuzzy partitions and, hence, to identify the corresponding consequent models which can be directly explained in terms of system behaviour presents a critical step in fuzzy neural modelling. In this paper, a new learning approach considering both nonlinear parameters in the rule premises and linear parameters in the rule consequents is proposed. Unlike the conventional two-stage optimization procedure widely practised in the field where the two sets of parameters are optimized separately, the consequent parameters are transformed into a dependent set on the premise parameters, thereby enabling the introduction of a new integrated gradient descent learning approach. A new Jacobian matrix is thus proposed and efficiently computed to achieve a more accurate approximation of the cost function by using the second-order Levenberg-Marquardt optimization method. Several other interpretability issues about the fuzzy neural model are also discussed and integrated into this new learning approach. Numerical examples are presented to illustrate the resultant structure of the fuzzy neural models and the effectiveness of the proposed new algorithm, and compared with the results from some well-known methods.

The failure of generalized gradient approximations (GGAs) and meta-GGAs for the two-center three-electron bonds in He-2(+), (H2O)(2)(+), and (NH3)(2)(+)

The radical cations He-2(+) (H2O)(2)(+), and (NH3)(2)(+) with two-center three-electron A-A bonds are investigated at the configuration interaction (CI), accurate Kohn-Sham (KS), generalized gradient approximation (GGA), and meta-GGA levels. Assessment of seven different GGA and six meta-GGA methods shows that the A(2)(+) systems remain a difficult case for density functional theory (DFT). All methods tested consistently overestimate the stability of A(2)(+): the corresponding D-e errors decrease for more diffuse valence densities in the series He-2(+) > (H2O)(2)(+) > (NH3)(2)(+). Upon comparison to the energy terms of the accurate Kohn-Sham solutions, the approximate exchange functionals are found to be responsible for the errors of GGA-type methods, which characteristically overestimate the exchange in A(2)(+). These so-called exchange functionals implicitly use localized holes. Such localized holes do occur if there is left-right correlation, i.e., the exchange functionals then also describe nondynamical correlation. However, in the hemibonded A(2)(+) systems the typical molecular (left-right, nondynamical) correlation of the two-electron pair bond is absent. The nondynamical correlation built into the exchange functionals is then spurious and yields too low energies.

Inorganic species of arsenic in soil solution determined by microcartridges and ferrihydrite-based diffusive gradient in thin films

The bioavailability of soil arsenic (As) is determined by its speciation in soil solution, i.e., arsenite [As(III)] or arsenate [As(V)]. Soil bioavailability studies require suitable methods to cope with small volumes of soil solution that can be speciated directly after sampling, and thereby minimise any As speciation change during sample collection. In this study, we tested a self-made microcartridge to separate both As species and compared it to a commercially available cartridge. In addition, the diffusive gradient in thin films technique (DGT), in combination with the microcartridges, was applied to synthetic solutions and to a soil spiked with As. This combination was used to improve the assessment of available inorganic As species with ferrihydrite(FH)-DGT, in order to validate the technique for environmental analysis, mainly in soils. The self-made microcartridge was effective in separating As(III) from As(V) in solution with detection by inductively coupled plasma optical emission spectrometry (ICP-OES) in volumes of only 3 ml. The DGT study also showed that the FH-based binding gels are effective for As(III) and As(V) assessment, in solutions with As and P concentrations and ionic strength commonly found in soils. The FH-DGT was tested on flooded and unflooded As spiked soils and recoveries of As(III) and As(V) were 85–104% of the total dissolved As. This study shows that the DGT with FH-based binding gel is robust for assessing inorganic species of As in soils.

Mean shift based gradient vector flow for image segmentation

In recent years, gradient vector flow (GVF) based algorithms have been successfully used to segment a variety of 2-D and 3-D imagery. However, due to the compromise of internal and external energy forces within the resulting partial differential equations, these methods may lead to biased segmentation results. In this paper, we propose MSGVF, a mean shift based GVF segmentation algorithm that can successfully locate the correct borders. MSGVF is developed so that when the contour reaches equilibrium, the various forces resulting from the different energy terms are balanced. In addition, the smoothness constraint of image pixels is kept so that over- or under-segmentation can be reduced. Experimental results on publicly accessible datasets of dermoscopic and optic disc images demonstrate that the proposed method effectively detects the borders of the objects of interest.