A Fast Algorithm for Sparse Support Vector Machines for Mobile Computing Applications

This research presents a fast algorithm for projected support vector machines (PSVM) by selecting a basis vector set (BVS) for the kernel-induced feature space, the training points are projected onto the subspace spanned by the selected BVS. A standard linear support vector machine (SVM) is then produced in the subspace with the projected training points. As the dimension of the subspace is determined by the size of the selected basis vector set, the size of the produced SVM expansion can be specified. A two-stage algorithm is derived which selects and refines the basis vector set achieving a locally optimal model. The model expansion coefficients and bias are updated recursively for increase and decrease in the basis set and support vector set. The condition for a point to be classed as outside the current basis vector and selected as a new basis vector is derived and embedded in the recursive procedure. This guarantees the linear independence of the produced basis set. The proposed algorithm is tested and compared with an existing sparse primal SVM (SpSVM) and a standard SVM (LibSVM) on seven public benchmark classification problems. Our new algorithm is designed for use in the application area of human activity recognition using smart devices and embedded sensors where their sometimes limited memory and processing resources must be exploited to the full and the more robust and accurate the classification the more satisfied the user. Experimental results demonstrate the effectiveness and efficiency of the proposed algorithm. This work builds upon a previously published algorithm specifically created for activity recognition within mobile applications for the EU Haptimap project [1]. The algorithms detailed in this paper are more memory and resource efficient making them suitable for use with bigger data sets and more easily trained SVMs.

Process monitoring approach using fast moving window PCA

This paper introduces a fast algorithm for moving window principal component analysis (MWPCA) which will adapt a principal component model. This incorporates the concept of recursive adaptation within a moving window to (i) adapt the mean and variance of the process variables, (ii) adapt the correlation matrix, and (iii) adjust the PCA model by recomputing the decomposition. This paper shows that the new algorithm is computationally faster than conventional moving window techniques, if the window size exceeds 3 times the number of variables, and is not affected by the window size. A further contribution is the introduction of an N-step-ahead horizon into the process monitoring. This implies that the PCA model, identified N-steps earlier, is used to analyze the current observation. For monitoring complex chemical systems, this work shows that the use of the horizon improves the ability to detect slowly developing drifts.

A fast method for fuzzy neural modeling and refinement

In the identification of complex dynamic systems using fuzzy neural networks, one of the main issues is the curse of dimensionality, which makes it difficult to retain a large number of system inputs or to consider a large number of fuzzy sets. Moreover, due to the correlations, not all possible network inputs or regression vectors in the network are necessary and adding them simply increases the model complexity and deteriorates the network generalisation performance. In this paper, the problem is solved by first proposing a fast algorithm for selection of network terms, and then introducing a refinement procedure to tackle the correlation issue. Simulation results show the efficacy of the method.

Low power field programmable gate array implementation of fast digital signal processing algorithms: characterisation and manipulation of data locality

Dynamic power consumption is very dependent on interconnect, so clever mapping of digital signal processing algorithms to parallelised realisations with data locality is vital. This is a particular problem for fast algorithm implementations where typically, designers will have sacrificed circuit structure for efficiency in software implementation. This study outlines an approach for reducing the dynamic power consumption of a class of fast algorithms by minimising the index space separation; this allows the generation of field programmable gate array (FPGA) implementations with reduced power consumption. It is shown how a 50% reduction in relative index space separation results in a measured power gain of 36 and 37% over a Cooley-Tukey Fast Fourier Transform (FFT)-based solution for both actual power measurements for a Xilinx Virtex-II FPGA implementation and circuit measurements for a Xilinx Virtex-5 implementation. The authors show the generality of the approach by applying it to a number of other fast algorithms namely the discrete cosine, the discrete Hartley and the Walsh-Hadamard transforms.

A sparse representation based fast detection method for surface defect detection of bottle caps

A practical machine-vision-based system is developed for fast detection of defects occurring on the surface of bottle caps. This system can be used to extract the circular region as the region of interests (ROI) from the surface of a bottle cap, and then use the circular region projection histogram (CRPH) as the matching features. We establish two dictionaries for the template and possible defect, respectively. Due to the requirements of high-speed production as well as detecting quality, a fast algorithm based on a sparse representation is proposed to speed up the searching. In the sparse representation, non-zero elements in the sparse factors indicate the defect's size and position. Experimental results in industrial trials show that the proposed method outperforms the orientation code method (OCM) and is able to produce promising results for detecting defects on the surface of bottle caps.

Part-whole relations between food webs and the validity of local food-web descriptions

This work analyzes the relationship between large food webs describing potential feeding relations between species and smaller sub-webs thereof describing relations actually realized in local communities of various sizes. Special attention is given to the relationships between patterns of phylogenetic correlations encountered in large webs and sub-webs. Based on the current theory of food-web topology as implemented in the matching model, it is shown that food webs are scale invariant in the following sense: given a large web described by the model, a smaller, randomly sampled sub-web thereof is described by the model as well. A stochastic analysis of model steady states reveals that such a change in scale goes along with a re-normalization of model parameters. Explicit formulae for the renormalized parameters are derived. Thus, the topology of food webs at all scales follows the same patterns, and these can be revealed by data and models referring to the local scale alone. As a by-product of the theory, a fast algorithm is derived which yields sample food webs from the exact steady state of the matching model for a high-dimensional trophic niche space in finite time. (C) 2008 Elsevier B.V. All rights reserved.

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.

Sensitivity and specificity of frequency-doubling technology, tendency-oriented perimetry, and Humphrey Swedish interactive threshold algorithm-fast perimetry in a glaucoma practice

PURPOSE: To evaluate the sensitivity and specificity of the screening mode of the Humphrey-Welch Allyn frequency-doubling technology (FDT), Octopus tendency-oriented perimetry (TOP), and the Humphrey Swedish Interactive Threshold Algorithm (SITA)-fast (HSF) in patients with glaucoma. DESIGN: A comparative consecutive case series. METHODS: This was a prospective study which took place in the glaucoma unit of an academic department of ophthalmology. One eye of 70 consecutive glaucoma patients and 28 age-matched normal subjects was studied. Eyes were examined with the program C-20 of FDT, G1-TOP, and 24-2 HSF in one visit and in random order. The gold standard for glaucoma was presence of a typical glaucomatous optic disk appearance on stereoscopic examination, which was judged by a glaucoma expert. The sensitivity and specificity, positive and negative predictive value, and receiver operating characteristic (ROC) curves of two algorithms for the FDT screening test, two algorithms for TOP, and three algorithms for HSF, as defined before the start of this study, were evaluated. The time required for each test was also analyzed. RESULTS: Values for area under the ROC curve ranged from 82.5%-93.9%. The largest area (93.9%) under the ROC curve was obtained with the FDT criteria, defining abnormality as presence of at least one abnormal location. Mean test time was 1.08 ± 0.28 minutes, 2.31 ± 0.28 minutes, and 4.14 ± 0.57 minutes for the FDT, TOP, and HSF, respectively. The difference in testing time was statistically significant (P <.0001). CONCLUSIONS: The C-20 FDT, G1-TOP, and 24-2 HSF appear to be useful tools to diagnose glaucoma. The test C-20 FDT and G1-TOP take approximately 1/4 and 1/2 of the time taken by 24 to 2 HSF. © 2002 by Elsevier Science Inc. All rights reserved.

Fast Moving Window Algorithm for QR and Cholesky Decompositions

This paper proposes a fast moving window algorithm for QR and Cholesky decompositions by simultaneously applying data updating and downdating. The developed procedure is based on inner products and entails a similar downdating to that of the Chambers’ approach. For adding and deleting one row of data from the original matrix, a detailed analysis shows that the proposed algorithm outperforms existing ones in terms or computational efficiency, if the number of columns exceeds 7. For a large number of columns, the proposed algorithm is numerically superior compared to the traditional sequential technique.