Fast identification algorithms for Gaussian process model

A class identification algorithms is introduced for Gaussian process(GP)models.The fundamental approach is to propose a new kernel function which leads to a covariance matrix with low rank,a property that is consequently exploited for computational efficiency for both model parameter estimation and model predictions.The objective of either maximizing the marginal likelihood or the Kullback–Leibler (K–L) divergence between the estimated output probability density function(pdf)and the true pdf has been used as respective cost functions.For each cost function,an efficient coordinate descent algorithm is proposed to estimate the kernel parameters using a one dimensional derivative free search, and noise variance using a fast gradient descent algorithm. Numerical examples are included to demonstrate the effectiveness of the new identification approaches.

Elastic net orthogonal forward regression

An efficient two-level model identification method aiming at maximising a model׳s generalisation capability is proposed for a large class of linear-in-the-parameters models from the observational data. A new elastic net orthogonal forward regression (ENOFR) algorithm is employed at the lower level to carry out simultaneous model selection and elastic net parameter estimation. The two regularisation parameters in the elastic net are optimised using a particle swarm optimisation (PSO) algorithm at the upper level by minimising the leave one out (LOO) mean square error (LOOMSE). There are two elements of original contributions. Firstly an elastic net cost function is defined and applied based on orthogonal decomposition, which facilitates the automatic model structure selection process with no need of using a predetermined error tolerance to terminate the forward selection process. Secondly it is shown that the LOOMSE based on the resultant ENOFR models can be analytically computed without actually splitting the data set, and the associate computation cost is small due to the ENOFR procedure. Consequently a fully automated procedure is achieved without resort to any other validation data set for iterative model evaluation. Illustrative examples are included to demonstrate the effectiveness of the new approaches.

Comparing hybrid data assimilation methods on the Lorenz 1963 model with increasing nonlinearity

We systematically compare the performance of ETKF-4DVAR, 4DVAR-BEN and 4DENVAR with respect to two traditional methods (4DVAR and ETKF) and an ensemble transform Kalman smoother (ETKS) on the Lorenz 1963 model. We specifically investigated this performance with increasing nonlinearity and using a quasi-static variational assimilation algorithm as a comparison. Using the analysis root mean square error (RMSE) as a metric, these methods have been compared considering (1) assimilation window length and observation interval size and (2) ensemble size to investigate the influence of hybrid background error covariance matrices and nonlinearity on the performance of the methods. For short assimilation windows with close to linear dynamics, it has been shown that all hybrid methods show an improvement in RMSE compared to the traditional methods. For long assimilation window lengths in which nonlinear dynamics are substantial, the variational framework can have diffculties fnding the global minimum of the cost function, so we explore a quasi-static variational assimilation (QSVA) framework. Of the hybrid methods, it is seen that under certain parameters, hybrid methods which do not use a climatological background error covariance do not need QSVA to perform accurately. Generally, results show that the ETKS and hybrid methods that do not use a climatological background error covariance matrix with QSVA outperform all other methods due to the full flow dependency of the background error covariance matrix which also allows for the most nonlinearity.

Adding a treatment arm to an ongoing clinical trial: a review of methodology and practice

Incorporating an emerging therapy as a new randomisation arm in a clinical trial that is open to recruitment would be desirable to researchers, regulators and patients to ensure that the trial remains current, new treatments are evaluated as quickly as possible, and the time and cost for determining optimal therapies is minimised. It may take many years to run a clinical trial from concept to reporting within a rapidly changing drug development environment; hence, in order for trials to be most useful to inform policy and practice, it is advantageous for them to be able to adapt to emerging therapeutic developments. This paper reports a comprehensive literature review on methodologies for, and practical examples of, amending an ongoing clinical trial by adding a new treatment arm. Relevant methodological literature describing statistical considerations required when making this specific type of amendment is identified, and the key statistical concepts when planning the addition of a new treatment arm are extracted, assessed and summarised. For completeness, this includes an assessment of statistical recommendations within general adaptive design guidance documents. Examples of confirmatory ongoing trials designed within the frequentist framework that have added an arm in practice are reported; and the details of the amendment are reviewed. An assessment is made as to how well the relevant statistical considerations were addressed in practice, and the related implications. The literature review confirmed that there is currently no clear methodological guidance on this topic, but that guidance would be advantageous to help this efficient design amendment to be used more frequently and appropriately in practice. Eight confirmatory trials were identified to have added a treatment arm, suggesting that trials can benefit from this amendment and that it can be practically feasible; however, the trials were not always able to address the key statistical considerations, often leading to uninterpretable or invalid outcomes. If the statistical concepts identified within this review are considered and addressed during the design of a trial amendment, it is possible to effectively assess a new treatment arm within an ongoing trial without compromising the original trial outcomes.

Optimal power management strategy for energy storage with stochastic loads

In this paper, a power management strategy (PMS) has been developed for the control of energy storage in a system subjected to loads of random duration. The PMS minimises the costs associated with the energy consumption of specific systems powered by a primary energy source and equipped with energy storage, under the assumption that the statistical distribution of load durations is known. By including the variability of the load in the cost function, it was possible to define the optimality criteria for the power flow of the storage. Numerical calculations have been performed obtaining the control strategies associated with the global minimum in energy costs, for a wide range of initial conditions of the system. The results of the calculations have been tested on a MATLAB/Simulink model of a rubber tyre gantry (RTG) crane equipped with a flywheel energy storage system (FESS) and subjected to a test cycle, which corresponds to the real operation of a crane in the Port of Felixstowe. The results of the model show increased energy savings and reduced peak power demand with respect to existing control strategies, indicating considerable potential savings for port operators in terms of energy and maintenance costs.

Obfuscation and Strict Online Anonymity

I consider the case for genuinely anonymous web searching. Big data seems to have it in for privacy. The story is well known, particularly since the dawn of the web. Vastly more personal information, monumental and quotidian, is gathered than in the pre-digital days. Once gathered it can be aggregated and analyzed to produce rich portraits, which in turn permit unnerving prediction of our future behavior. The new information can then be shared widely, limiting prospects and threatening autonomy. How should we respond? Following Nissenbaum (2011) and Brunton and Nissenbaum (2011 and 2013), I will argue that the proposed solutions—consent, anonymity as conventionally practiced, corporate best practices, and law—fail to protect us against routine surveillance of our online behavior. Brunton and Nissenbaum rightly maintain that, given the power imbalance between data holders and data subjects, obfuscation of one’s online activities is justified. Obfuscation works by generating “misleading, false, or ambiguous data with the intention of confusing an adversary or simply adding to the time or cost of separating good data from bad,” thus decreasing the value of the data collected (Brunton and Nissenbaum, 2011). The phenomenon is as old as the hills. Natural selection evidently blundered upon the tactic long ago. Take a savory butterfly whose markings mimic those of a toxic cousin. From the point of view of a would-be predator the data conveyed by the pattern is ambiguous. Is the bug lunch or potential last meal? In the light of the steep costs of a mistake, the savvy predator goes hungry. Online obfuscation works similarly, attempting for instance to disguise the surfer’s identity (Tor) or the nature of her queries (Howe and Nissenbaum 2009). Yet online obfuscation comes with significant social costs. First, it implies free riding. If I’ve installed an effective obfuscating program, I’m enjoying the benefits of an apparently free internet without paying the costs of surveillance, which are shifted entirely onto non-obfuscators. Second, it permits sketchy actors, from child pornographers to fraudsters, to operate with near impunity. Third, online merchants could plausibly claim that, when we shop online, surveillance is the price we pay for convenience. If we don’t like it, we should take our business to the local brick-and-mortar and pay with cash. Brunton and Nissenbaum have not fully addressed the last two costs. Nevertheless, I think the strict defender of online anonymity can meet these objections. Regarding the third, the future doesn’t bode well for offline shopping. Consider music and books. Intrepid shoppers can still find most of what they want in a book or record store. Soon, though, this will probably not be the case. And then there are those who, for perfectly good reasons, are sensitive about doing some of their shopping in person, perhaps because of their weight or sexual tastes. I argue that consumers should not have to pay the price of surveillance every time they want to buy that catchy new hit, that New York Times bestseller, or a sex toy.

The effects of attitudes and personality traits on mode choice

We hypothesise that differences in people's attitudes and personality traits lead them to attribute varying importance to environmental considerations, safety, comfort, convenience and flexibility. Differences in personality traits call be revealed not only in the individuals' choice of transport, but also in other actions of their everyday lives-such as how much they recycle, whether they take precautions or avoid dangerous pursuits. Conditioning on a set of exogenous individual characteristics, we use indicators of attitudes and personality traits to form latent variables for inclusion in an, otherwise standard, discrete mode choice model. With a sample of Swedish commuters, we find that both attitudes towards flexibility and comfort, as well as being pro-environmentally inclined, influence the individual's choice of mode. Although modal time and cost still are important, it follows that there are other ways, apart from economic incentives, to attract individuals to the, from society's perspective, desirable public modes of transport. Our results should provide useful information to policy-makers and transportation planners developing sustainable transportation systems.

Ubiquitous computing in education : a SWOT analysis by students and teachers

Learning from anywhere anytime is a contemporary phenomenon in the field of education that is thought to be flexible, time and cost saving. The phenomenon is evident in the way computer technology mediates knowledge processes among learners. Computer technology is however, in some instances, faulted. There are studies that highlight drawbacks of computer technology use in learning. In this study we aimed at conducting a SWOT analysis on ubiquitous computing and computer-mediated social interaction and their affect on education. Students and teachers were interviewed on the mentioned concepts using focus group interviews. Our contribution in this study is, identifying what teachers and students perceive to be the strength, weaknesses, opportunities and threats of ubiquitous computing and computer-mediated social interaction in education. We also relate the findings with literature and present a common understanding on the SWOT of these concepts. Results show positive perceptions. Respondents revealed that ubiquitous computing and computer-mediated social interaction are important in their education due to advantages such as flexibility, efficiency in terms of cost and time, ability to acquire computer skills. Nevertheless disadvantages where also mentioned for example health effects, privacy and security issues, noise in the learning environment, to mention but a few. This paper gives suggestions on how to overcome threats mentioned.

Spatial prediction of rufous bristlebird habitat in a coastal heathland: a GIS-based approach

1. To develop a conservation management plan for a species, knowledge of its distribution and spatial arrangement of preferred habitat is essential. This is a difficult task, especially when the species of concern is in low   abundance. In south-western Victoria, Australia, populations of the rare rufous bristlebird Dasyornis broadbenti are threatened by fragmentation of suitable habitat. In order to improve the conservation status of this species, critical habitat requirements must be identified and a system of corridors must be established to link known populations. A predictive spatial model of rufous bristlebird habitat was developed in order to identify critical areas requiring preservation, such as corridors for dispersal.
2. Habitat models generated using generalized linear modelling techniques can assist in delineating the specific habitat requirements of a species. Coupled with geographic information system (GIS) technology, these models can be extrapolated to produce maps displaying the spatial configuration of suitable habitat.
3. Models were generated using logistic regression, with bristlebird presence or absence as the dependent variable and landscape variables, extracted from both GIS data layers and multispectral digital imagery, as the predictors. A multimodel inference approach based on Akaike’s information criterion was used and the resulting model was applied in a GIS to extrapolate predicted likelihood of occurrence across the entire area of concern. The predictive performance of the selected model was evaluated using the receiver operating characteristic (ROC) technique. A hierarchical partitioning protocol was used to identify the predictor variables most likely to influence variation in the dependent variable. Probability of species presence was used as an index of habitat suitability.
4. Negative associations between rufous bristlebird presence and  increasing elevation, 'distance to cree', 'distance to coast' and sun index were evident, suggesting a preference for areas relatively low in altitude, in close proximity to the coastal fringe and drainage lines, and receiving less direct sunlight. A positive association with increasing habitat complexity also suggested that this species prefers areas containing high vertical density of vegetation.
5. The predictive performance of the selected model was shown to be high (area under the curve 0·97), indicating a good fit of the model to the data. Hierarchical partitioning analysis showed that all the variables considered had significant  independent contributions towards explaining the variation in the dependent variable. The proportion of the total study area that was predicted as suitable habitat for the rufous bristlebird (using probability of occurrence at a ≥0·5 level ) was 16%.
6. Synthesis and applications. The spatial model clearly delineated areas predicted as highly suitable rufous bristlebird habitat, with evidence of potential corridors linking coastal and inland populations via gullies. Conservation of this species will depend on management actions that protect the critical habitats identified in the model. A multi-scale  approach to the modelling process is recommended whereby a spatially explicit model is first generated using landscape variables extracted from a GIS, and a second model at site level is developed using fine-scale habitat variables measured on the ground. Where there are constraints on the time and cost involved in measuring finer scale variables, the first step alone can be used for conservation planning.

Perceived personal, social and environmental barriers to weight maintenance among young women: a community survey

Background: Young women are a group at high risk of weight gain. This study examined a range of perceived personal, social and environmental barriers to physical activity and healthy eating for weight maintenance among young women, and how these varied by socioeconomic status (SES), overweight status and domestic situation. Methods: In October-December 2001, a total of 445 women aged 18–32 years, selected randomly from the Australian electoral roll, completed a mailed self-report survey that included questions on 11 barriers to physical activity and 11 barriers to healthy eating (relating to personal, social and environmental factors). Height, weight and socio-demographic details were also obtained. Statistical analyses were conducted mid-2003. Results: The most common perceived barriers to physical activity and healthy eating encountered by young women were related to motivation, time and cost. Women with children were particularly likely to report a lack of social support as an important barrier to physical activity, and lack of social support and time as important barriers to healthy eating. Perceived barriers did not differ by SES or overweight status. Conclusions: Health promotion strategies aimed at preventing weight gain should take into account the specific perceived barriers to physical activity and healthy eating faced by women in this age group, particularly lack of motivation, lack of time, and cost. Strategies targeting perceived lack of time and lack of social support are particularly required for young women with children.

A new second-order method for blind signal separation from dynamic mixtures

This paper presents a new approach to separate colored stationary signals mixed by convolutive channels. A cost function is proposed by employing linear constraint to the demixing vectors. The linear constraint is shown to be sufficient for avoiding trivial solution. The minimization of the cost function is performed using the Lagrangian method. Simulation results demonstrate the performance of the algorithm.

A new second order method for blind signal separation from convolutive mixtures

This paper presents a new approach to separate colored signals mixed by FIR (finite impulse response) and MIMO (multiple-input multiple-output) channels. A cost function is proposed by employing linear constrainit to the de mixing vectors. The linear constraint is shown to be sufficient for avoiding trivial solution. The minimization of the cost function is performed using the Lagrangian method. Simulation results demonstrate the performance of the algorithm.

A modified CM algorithm for blind equalization of MPSK signals

A new blind equalization algorithm for application to wireless communication employing MPSK signals is proposed in this paper.  Since the new cost function exploits the amplitude and phase information simultaneously, the proposed algorithm can provide a superior performance than the conventional constant modulus algorithm (CMA) which only use the amplitude knowledge in its cost function.  Theoretical analysis and numerical simulations both demonstrate that the steady-state mean square error (MSE) for the proposed algorithm is less than that of the CMA.

On the effect of path loss rates on the optimal receiver position in sensor networks

In this paper we examine the problem of finding an optimal position for a receiver node in a single-hop sensor network. The basic idea is to minimize the network energy consumption according to a particular network cost function. Our contribution is to simply show the effect of signal path loss rates and sensor weighting on the optimal receiver position.

The effect of osteoporotic vertebral fracture of predicted spinal loads in vivo

he aetiology of osteoporotic vertebral fractures is multi-factorial, and cannot be explained solely by low bone mass. After sustaining an initial vertebral fracture, the risk of subsequent fracture increases greatly. Examination of physiologic loads imposed on vertebral bodies may help to explain a mechanism underlying this fracture cascade. This study tested the hypothesis that model-derived segmental vertebral loading is greater in individuals who have sustained an osteoporotic vertebral fracture compared to those with osteoporosis and no history of fracture. Flexion moments, and compression and shear loads were calculated from T2 to L5 in 12 participants with fractures (66.4 ± 6.4 years, 162.2 ± 5.1 cm, 69.1 ± 11.2 kg) and 19 without fractures (62.9 ± 7.9 years, 158.3 ± 4.4 cm, 59.3 ± 8.9 kg) while standing. Static analysis was used to solve gravitational loads while muscle-derived forces were calculated using a detailed trunk muscle model driven by optimization with a cost function set to minimise muscle fatigue. Least squares regression was used to derive polynomial functions to describe normalised load profiles. Regression co-efficients were compared between groups to examine differences in loading profiles. Loading at the fractured level, and at one level above and below, were also compared between groups. The fracture group had significantly greater normalised compression (p = 0.0008) and shear force (p < 0.0001) profiles and a trend for a greater flexion moment profile. At the level of fracture, a significantly greater flexion moment (p = 0.001) and shear force (p < 0.001) was observed in the fracture group. A greater flexion moment (p = 0.003) and compression force (p = 0.007) one level below the fracture, and a greater flexion moment (p = 0.002) and shear force (p = 0.002) one level above the fracture was observed in the fracture group. The differences observed in multi-level spinal loading between the groups may explain a mechanism for increased risk of subsequent vertebral fractures. Interventions aimed at restoring vertebral morphology or reduce thoracic curvature may assist in normalising spine load profiles.