Dictionary learning and sparse coding on Grassmann manifolds : an extrinsic solution

Recent advances in computer vision and machine learning suggest that a wide range of problems can be addressed more appropriately by considering non-Euclidean geometry. In this paper we explore sparse dictionary learning over the space of linear subspaces, which form Riemannian structures known as Grassmann manifolds. To this end, we propose to embed Grassmann manifolds into the space of symmetric matrices by an isometric mapping, which enables us to devise a closed-form solution for updating a Grassmann dictionary, atom by atom. Furthermore, to handle non-linearity in data, we propose a kernelised version of the dictionary learning algorithm. Experiments on several classification tasks (face recognition, action recognition, dynamic texture classification) show that the proposed approach achieves considerable improvements in discrimination accuracy, in comparison to state-of-the-art methods such as kernelised Affine Hull Method and graph-embedding Grassmann discriminant analysis.

Generation of two isomers of C5H from the corresponding anions. A theoretically motivated mass spectrometric study

Molecular orbital calculations have predicted the stability of a range of connectivities for the radical C5H potential surface. The most energetically favorable of these include the linear C4CH geometry and two ring-chain structures HC2C3 and C2C3H The corresponding anions are also shown to be theoretically stable, and furthermore, a fourth isomer, C2CHC2, is predicted to be the most stable anion connectivity. These results have motivated experimental efforts. Methodologies for the generation of the non-ring-containing isomeric anions C4CH and C2CHC2 have been developed utilizing negative ion mass spectrometry. The absolute connectivities of the anions have been established using deuterium labeling, charge reversal, and neutralization reionization techniques. The success of the latter experiment confirms theoretical predictions of stability of the corresponding neutral species. This is the first reported observation of the neutral C2CHC2 species that calculations predict to be substantially less stable than the C4CH connectivity but still bound relative to isomerization processes.

Activation of matrix metalloproteinase-2 (MMP-2) by membrane type 1 matrix metalloproteinase through an artificial receptor for ProMMP-2 generates active MMP-2

The suggested model for pro-matrix metalloproteinase-2 (proMMP-2) activation by membrane type 1 MMP (MT1-MMP) implicates the complex between MT1-MMP and tissue inhibitor of MMP-2 (TIMP-2) as a receptor for proMMP-2. To dissect this model and assess the pathologic significance of MMP-2 activation, an artificial receptor for proMMP-2 was created by replacing the signal sequence of TIMP-2 with cytoplasmic/transmembrane domain of type II transmembrane mosaic serine protease (MSP-T2). Unlike TIMP-2, MSP-T2 served as a receptor for proMMP-2 without inhibiting MT1-MMP, and generated TIMP-2-free active MMP-2 even at a low level of MT1-MMP. Thus, MSP-T2 did not affect direct cleavage of the substrate testican-1 by MT1-MMP, whereas TIMP-2 inhibited it even at the level that stimulates proMMP-2 processing. Expression of MSP-T2 in HT1080 cells enhanced MMP-2 activation by endogenous MT1-MMP and caused intensive hydrolysis of collagen gel. Expression of MSP-T2 in U87 glioma cells, which express a trace level of endogenous MT1-MMP, induced MMP-2 activation and enhanced cell-associated protease activity, activation of extracellular signal-regulated kinase, and metastatic ability into chick embryonic liver and lung. MT1-MMP can exert both maximum MMP-2 activation and direct cleavage of substrates with MSP-T2, which cannot be achieved with TIMP-2. These results suggest that MMP-2 activation by MT1-MMP potentially amplifies protease activity, and combination with direct cleavage of substrate causes effective tissue degradation and enhances tumor invasion and metastasis, which highlights the complex role of TIMP-2. MSP-T2 is a unique tool to analyze physiologic and pathologic roles of MMP-2 and MT1-MMP in comparison with TIMP-2.

Artificial neural networks to predict activity type and energy expenditure in youth

Previous studies have demonstrated that pattern recognition approaches to accelerometer data reduction are feasible and moderately accurate in classifying activity type in children. Whether pattern recognition techniques can be used to provide valid estimates of physical activity (PA) energy expenditure in youth remains unexplored in the research literature. Purpose: The objective of this study is to develop and test artificial neural networks (ANNs) to predict PA type and energy expenditure (PAEE) from processed accelerometer data collected in children and adolescents. Methods: One hundred participants between the ages of 5 and 15 yr completed 12 activity trials that were categorized into five PA types: sedentary, walking, running, light-intensity household activities or games, and moderate-to-vigorous intensity games or sports. During each trial, participants wore an ActiGraph GTIM on the right hip, and (V) Over dotO(2) was measured using the Oxycon Mobile (Viasys Healthcare, Yorba Linda, CA) portable metabolic system. ANNs to predict PA type and PAEE (METs) were developed using the following features: 10th, 25th, 50th, 75th, and 90th percentiles and the lag one autocorrelation. To determine the highest time resolution achievable, we extracted features from 10-, 15-, 20-, 30-, and 60-s windows. Accuracy was assessed by calculating the percentage of windows correctly classified and root mean square en-or (RMSE). Results: As window size increased from 10 to 60 s, accuracy for the PA-type ANN increased from 81.3% to 88.4%. RMSE for the MET prediction ANN decreased from 1.1 METs to 0.9 METs. At any given window size, RMSE values for the MET prediction ANN were 30-40% lower than the conventional regression-based approaches. Conclusions: ANNs can be used to predict both PA type and PAEE in children and adolescents using count data from a single waist mounted accelerometer.

Interpretation of association rules in multi-tier structures

Dealing with the large amount of data resulting from association rule mining is a big challenge. The essential issue is how to provide efficient methods for summarizing and representing meaningful discovered knowledge from databases. This paper presents a new approach called multi-tier granule mining to improve the performance of association rule mining. Rather than using patterns, it uses granules to represent knowledge that is implicitly contained in relational databases. This approach also uses multi-tier structures and association mappings to interpret association rules in terms of granules. Consequently, association rules can be quickly assessed and meaningless association rules can be justified according to these association mappings. The experimental results indicate that the proposed approach is promising

Non-linear model predictive control for UAVs with slung/swung load

Suspended loads on UAVs can provide significant benefits to several applications in agriculture, law enforcement and construction. The load impact on the underlying system dynamics should not be neglected as significant feedback forces may be induced on the vehicle during certain flight manoeuvres. Much research has focused on standard multi-rotor position and attitude control with and without a slung load. However, predictive control schemes, such as Nonlinear Model Predictive Control (NMPC), have not yet been fully explored. To this end, we present software and flight system architecture to test controller for safe and precise operation of multi-rotors with heavy slung load in three dimensions.

Non-linear principal components analysis : an alternative method for finding patterns in environmental data

The main purpose of this article is to gain an insight into the relationships between variables describing the environmental conditions of the Far Northern section of the Great Barrier Reef, Australia. Several of the variables describing these conditions had different measurement levels and often they had non-linear relationships. Using non-linear principal component analysis, it was possible to acquire an insight into these relationships. Furthermore, three geographical areas with unique environmental characteristics could be identified.

Benthic metabolism as an indicator of stream ecosystem health

We tested direct and indirect measures of benthic metabolism as indicators of stream ecosystem health across a known agricultural land-use disturbance gradient in southeast Queensland, Australia. Gross primary production (GPP) and respiration (R24) in benthic chambers in cobble and sediment habitats, algal biomass (as chlorophyll a) from cobbles and sediment cores, algal biomass accrual on artificial substrates and stable carbon isotope ratios of aquatic plants and benthic sediments were measured at 53 stream sites, ranging from undisturbed subtropical rainforest to catchments where improved pasture and intensive cropping are major land-uses. Rates of benthic GPP and R24 varied by more than two orders of magnitude across the study gradient. Generalised linear regression modelling explained 80% or more of the variation in these two indicators when sediment and cobble substrate dominated sites were considered separately, and both catchment and reach scale descriptors of the disturbance gradient were important in explaining this variation. Model fits were poor for net daily benthic metabolism (NDM) and production to respiration ratio (P/R). Algal biomass accrual on artificial substrate and stable carbon isotope ratios of aquatic plants and benthic sediment were the best of the indirect indicators, with regression model R2 values of 50% or greater. Model fits were poor for algal biomass on natural substrates for cobble sites and all sites. None of these indirect measures of benthic metabolism was a good surrogate for measured GPP. Direct measures of benthic metabolism, GPP and R24, and several indirect measures were good indicators of stream ecosystem health and are recommended in assessing process-related responses to riparian and catchment land use change and the success of ecosystem rehabilitation actions.

Cheating prevention in linear secret sharing

Cheating detection in linear secret sharing is considered. The model of cheating extends the Tompa-Woll attack and includes cheating during multiple (unsuccessful) recovery of the secret. It is shown that shares in most linear schemes can be split into subshares. Subshares can be used by participants to trade perfectness of the scheme with cheating prevention. Evaluation of cheating prevention is given in the context of different strategies applied by cheaters.

Distinguishing attack on SOBER-128 with linear masking

We present a distinguishing attack against SOBER-128 with linear masking. We found a linear approximation which has a bias of 2^− − 8.8 for the non-linear filter. The attack applies the observation made by Ekdahl and Johansson that there is a sequence of clocks for which the linear combination of some states vanishes. This linear dependency allows that the linear masking method can be applied. We also show that the bias of the distinguisher can be improved (or estimated more precisely) by considering quadratic terms of the approximation. The probability bias of the quadratic approximation used in the distinguisher is estimated to be equal to O(2^− − 51.8), so that we claim that SOBER-128 is distinguishable from truly random cipher by observing O(2^103.6) keystream words.

On governance structures for the cloud computing services and assessing their effectiveness

This research suggests information technology (IT) governance structures to manage the cloud computing services. The interest in acquiring IT resources as a utility from the cloud computing environment is gaining momentum. The cloud computing services present organizations with opportunities to manage their IT expenditure on an ongoing basis, and access to modern IT resources to innovate and manage their continuity. However, the cloud computing services are no silver bullet. Organizations would need to have appropriate governance structures and policies in place to manage the cloud computing services. The subsequent decisions from these governance structures will ensure the effective management of the cloud computing services. This management will facilitate a better fit of the cloud computing services into organizations’ existing processes to achieve the business (process-level) and the financial (firm-level) objectives. Using a triangulation approach, we suggest four governance structures for managing the cloud computing services. These structures are a chief cloud officer, a cloud management committee, a cloud service facilitation centre, and a cloud relationship centre. We also propose that these governance structures would relate directly to organizations cloud computing services-related business objectives, and indirectly to cloud computing services-related financial objectives. Perceptive field survey data from actual and prospective cloud computing service adopters suggest that the suggested governance structures would contribute directly to cloud computing-related business objectives and indirectly to cloud computing-related financial objectives.

Solitary filamentary structures and nanosecond dynamics in atmospheric-pressure plasmas driven by tailored dc pulses

Nanosecond dynamics of two separated discharge cycles in an asymmetric dielectric barrier discharge is studied using time-resolved current and voltage measurements synchronized with high-speed (∼5 ns) optical imaging. Nanosecond dc pulses with tailored raise and fall times are used to generate solitary filamentary structures (SFSs) during the first cycle and a uniform glow during the second. The SFSs feature ∼1.5 mm thickness, ∼1.9 A peak current, and a lifetime of several hundred nanoseconds, at least an order of magnitude larger than in common microdischarges. This can be used in alternating localized and uniform high-current plasma treatments in various applications.

Integrated plasma-aided nanofabrication facility: Operation, parameters, and assembly of quantum structures and functional nanomaterials

The development, operation, and applications of two configurations of an integrated plasma-aided nanofabrication facility (IPANF) comprising low-frequency inductively coupled plasma-assisted, low-pressure, multiple-target RF magnetron sputtering plasma source, are reported. The two configurations of the plasma source have different arrangements of the RF inductive coil: a conventional external flat spiral "pancake" coil and an in-house developed internal antenna comprising two orthogonal RF current sheets. The internal antenna configuration generates a "unidirectional" RF current that deeply penetrates into the plasma bulk and results in an excellent uniformity of the plasma over large areas and volumes. The IPANF has been employed for various applications, including low-temperature plasma-enhanced chemical vapor deposition of vertically aligned single-crystalline carbon nanotips, growth of ultra-high aspect ratio semiconductor nanowires, assembly of optoelectronically important Si, SiC, and Al1-xInxN quantum dots, and plasma-based synthesis of bioactive hydroxyapatite for orthopedic implants.

Microscopic ion fluxes in plasma-aided nanofabrication of ordered carbon nanotip structures

Three-dimensional topography of microscopic ion fluxes in the reactive hydrocarbon-based plasma-aided nanofabrication of ordered arrays of vertically aligned single-crystalline carbon nanotip microemitter structures is simulated by using a Monte Carlo technique. The individual ion trajectories are computed by integrating the ion equations of motion in the electrostatic field created by a biased nanostructured substrate. It is shown that the ion flux focusing onto carbon nanotips is more efficient under the conditions of low potential drop Us across the near-substrate plasma sheath. Under low- Us conditions, the ion current density onto the surface of individual nanotips is higher for higher-aspect-ratio nanotips and can exceed the mean ion current density onto the entire nanopattern in up to approximately five times. This effect becomes less pronounced with increasing the substrate bias, with the mean relative enhancement of the ion current density ξi not exceeding ∼1.7. The value of ξi is higher in denser plasmas and behaves differently with the electron temperature Te depending on the substrate bias. When the substrate bias is low, ξi decreases with Te, with the opposite tendency under higher- Us conditions. The results are relevant to the plasma-enhanced chemical-vapor deposition of ordered large-area nanopatterns of vertically aligned carbon nanotips, nanofibers, and nanopyramidal microemitter structures for flat-panel display applications. © 2005 American Institute of Physics.

Unconventional Ni–P alloy-catalyzed CVD of carbon coil-like micro- and nano-structures

Conventional catalyzed thermal CVD of carbon microcoils commonly suffers from poor control of the coil shape and morphology and rarely reaches the nanoscale size range. This article reports on an unconventional Ni-P alloy-catalyzed, high-throughput, highly reproducible CVD of ultra-long carbon coil-like micro- and nano-structures using acetylene precursor at relatively low process temperatures. Helical carbon microcoils with consistently uniform, circular cross-sections and a high degree of crystallinity have been synthesized at 750 °C. A further reduction of the temperature to 650 °C led to the growth of ultra-long (up to several mm) wave-like carbon nanofibers made of two nanowires with the diameters in the 100-200 nm range. The results of the XRD and Raman analysis reveal that the nanofibers feature only a slightly more disordered structure compared to the microcoils. Our results suggest that morphology and structure of the carbon coil-like micro- and nano-structures can be tailored by the appropriate alloying of the catalyst and the choice of the CVD process parameters.