Particle swarm optimization aided orthogonal forward regression for unified data modeling

We propose a unified data modeling approach that is equally applicable to supervised regression and classification applications, as well as to unsupervised probability density function estimation. A particle swarm optimization (PSO) aided orthogonal forward regression (OFR) algorithm based on leave-one-out (LOO) criteria is developed to construct parsimonious radial basis function (RBF) networks with tunable nodes. Each stage of the construction process determines the center vector and diagonal covariance matrix of one RBF node by minimizing the LOO statistics. For regression applications, the LOO criterion is chosen to be the LOO mean square error, while the LOO misclassification rate is adopted in two-class classification applications. By adopting the Parzen window estimate as the desired response, the unsupervised density estimation problem is transformed into a constrained regression problem. This PSO aided OFR algorithm for tunable-node RBF networks is capable of constructing very parsimonious RBF models that generalize well, and our analysis and experimental results demonstrate that the algorithm is computationally even simpler than the efficient regularization assisted orthogonal least square algorithm based on LOO criteria for selecting fixed-node RBF models. Another significant advantage of the proposed learning procedure is that it does not have learning hyperparameters that have to be tuned using costly cross validation. The effectiveness of the proposed PSO aided OFR construction procedure is illustrated using several examples taken from regression and classification, as well as density estimation applications.

Probability density estimation with tunable kernels using orthogonal forward regression

A generalized or tunable-kernel model is proposed for probability density function estimation based on an orthogonal forward regression procedure. Each stage of the density estimation process determines a tunable kernel, namely, its center vector and diagonal covariance matrix, by minimizing a leave-one-out test criterion. The kernel mixing weights of the constructed sparse density estimate are finally updated using the multiplicative nonnegative quadratic programming algorithm to ensure the nonnegative and unity constraints, and this weight-updating process additionally has the desired ability to further reduce the model size. The proposed tunable-kernel model has advantages, in terms of model generalization capability and model sparsity, over the standard fixed-kernel model that restricts kernel centers to the training data points and employs a single common kernel variance for every kernel. On the other hand, it does not optimize all the model parameters together and thus avoids the problems of high-dimensional ill-conditioned nonlinear optimization associated with the conventional finite mixture model. Several examples are included to demonstrate the ability of the proposed novel tunable-kernel model to effectively construct a very compact density estimate accurately.

Radial basis function classifier construction using particle swarm optimisation aided orthogonal forward regression

We develop a particle swarm optimisation (PSO) aided orthogonal forward regression (OFR) approach for constructing radial basis function (RBF) classifiers with tunable nodes. At each stage of the OFR construction process, the centre vector and diagonal covariance matrix of one RBF node is determined efficiently by minimising the leave-one-out (LOO) misclassification rate (MR) using a PSO algorithm. Compared with the state-of-the-art regularisation assisted orthogonal least square algorithm based on the LOO MR for selecting fixednode RBF classifiers, the proposed PSO aided OFR algorithm for constructing tunable-node RBF classifiers offers significant advantages in terms of better generalisation performance and smaller model size as well as imposes lower computational complexity in classifier construction process. Moreover, the proposed algorithm does not have any hyperparameter that requires costly tuning based on cross validation.

Effect of feed restriction and supplemental dietary fat on gut peptide and hypothalamic neuropeptide mRNA concentrations in growing wethers

The objectives of the present study were 1) to evaluate the effects of supplemental fat and ME intake on plasma concentrations of glucagon-like peptide-1 (GLP-1), cholecystokinin (CCK), glucose-dependent insulinotropic polypeptide, ghrelin, and oxyntomodulin; and 2) to determine the association of these peptides with DMI and the hypothalamic concentration of mRNA for the following neuropeptides: neuropeptide Y (NPY), agouti-related peptide (AgRP), and proopiomelanocortin (POMC). In a completely randomized block design with a 2 x 2 factorial arrangement of treatments, 32 pens with 2 wethers each were restricted-fed (2.45 Mcal/lamb per day) or offered diets ad libitum (n = 16) with or without 6% supplemental fat (n = 16) for a period of 30 d. Dry matter intake was measured daily. On d 8, 15, 22, and 29, BW was measured before feeding, and 6 h after feeding, blood samples were collected for plasma measurement of insulin, GLP-1, CCK, ghrelin, glucose-dependent insulinotropic polypeptide, oxyntomodulin, glucose, and NEFA concentrations. On d 29, blood was collected 30 min before feeding for the same hormone and metabolite analyses. At the end of the experiment, wethers were slaughtered and the hypothalami were collected to measure concentrations of NPY, AgRP, and POMC mRNA. Offering feed ad libitum (resulting in greater ME intake) increased plasma insulin and NEFA concentrations (P = 0.02 and 0.02, respectively) and decreased hypothalamic mRNA expression of NPY and AgRP (P = 0.07 and 0.02, respectively) compared with the restricted-fed wethers. There was a trend for the addition of dietary fat to decrease DMI (P = 0.12). Addition of dietary fat decreased insulin and glucose concentrations (P < 0.05 and 0.01, respectively) and tended to increase hypothalamic mRNA concentrations for NPY and AgRP (P = 0.07 and 0.11, respectively). Plasma GLP-1 and CCK concentrations increased in wethers offered feed ad libitum compared with restricted-fed wethers, but the response was greater when wethers were offered feed ad libitum and had supplemental fat in the diet (fat x intake interaction, P = 0.04). The prefeeding plasma ghrelin concentration was greater in restricted-fed wethers compared with those offered feed ad libitum, but the concentrations were similar 6 h after feeding (intake x time interaction, P < 0.01). Supplemental dietary fat did not affect (P = 0.22) plasma ghrelin concentration. We conclude that insulin, ghrelin, CCK, and GLP-1 may regulate DMI in sheep by regulating the hypothalamic gene expression of NPY, AgRP, and POMC.

Proposing a way forward: a review of standardisation in the use of age categories and ageing techniques in osteological analysis (2004 to 2009)

This study assesses the current state of adult skeletal age-at-death estimation in biological anthropology through analysis of data published in recent research articles from three major anthropological and archaeological journals (2004–2009). The most commonly used adult ageing methods, age of ‘adulthood’, age ranges and the maximum age reported for ‘mature’ adults were compared. The results showed a wide range of variability in the age at which individuals were determined to be adult (from 14 to 25 years), uneven age ranges, a lack of standardisation in the use of descriptive age categories and the inappropriate application of some ageing methods for the sample being examined. Such discrepancies make comparisons between skeletal samples difficult, while the inappropriate use of some techniques make the resultant age estimations unreliable. At a time when national and even global comparisons of past health are becoming prominent, standardisation in the terminology and age categories used to define adults within each sample is fundamental. It is hoped that this research will prompt discussions in the osteological community (both nationally and internationally) about what defines an ‘adult’, how to standardise the age ranges that we use and how individuals should be assigned to each age category. Skeletal markers have been proposed to help physically identify ‘adult’ individuals.

A controlled trial of sip-feed supplements in elderly orthopaedic patients

The potential nutritional and clinical benefits of sip-feed supplements were investigated by means of a controlled trial in elderly female patients admitted for orthopaedic surgery. A nutritional risk assessment procedure (Nutritional Risk Questionnaire, NRQ) was used to identify patients who might benefit from supplementation. Patients identified as high risk who did not receive supplements showed significant losses in triceps skinfold thickness (TSF) and mid-upper arm muscle circumference (MUAMC) measurements during hospitalization. Such changes were not observed in high-risk supplemented patients, but significant losses of MUAMC were also recorded in a group of patients who failed to comply with the supplement. No differences in biochemical parameters, muscle function, or clinical outcome were observed between supplemented and unsupplemented and non-compliant patients. The problems of poor compliance to sip-feed supplements and failure to observe clinical benefit in supplemented patients are discussed.

Moving feedback forward: theory to practice

There is substantial research interest in tutor feedback and students’ perception and use of such feedback. This paper considers some of the major issues raised in relation to tutor feedback and student learning. We explore some of the current feedback drivers, most notably the need for feedback to move away from simply a monologue from a tutor to a student to a valuable tutor–student dialogue. In relation to moving feedback forward the notions of self regulation, dialogue and social learning are explored and then considered in relation to how such theory can translate into practice. The paper proposes a framework (GOALS) as a tool through which tutors can move theory into practice with the aim of improving student learning from feedback.

Forward pricing and the housing market: the pre-sales housing system in Taiwan

In countries that have experienced rapid economic development, the need to establish more efficient markets in which private property can be constructed has induced some innovative solutions. One such solution is the phenomenon of a pre-sales market of the kind that can be observed in Taiwan, Korea, and more recently in China. Developers sell their property before building is started in order to acquire financing for the development companies. This paper discusses the process and, by recognising the analogy between the pre-sales market and forwards markets, analyses the implications for developers