Fast cross-validation of kernel fisher discriminant classifiers

Given n training examples, the training of a Kernel Fisher Discriminant (KFD) classifier corresponds to solving a linear system of dimension n. In cross-validating KFD, the training examples are split into 2 distinct subsets for a number of times (L) wherein a subset of m examples is used for validation and the other subset of(n - m) examples is used for training the classifier. In this case L linear systems of dimension (n - m) need to be solved. We propose a novel method for cross-validation of KFD in which instead of solving L linear systems of dimension (n - m), we compute the inverse of an n × n matrix and solve L linear systems of dimension 2m, thereby reducing the complexity when L is large and/or m is small. For typical 10-fold and leave-one-out cross-validations, the proposed algorithm is approximately 4 and (^4/9n) times respectively as efficient as the naive implementations. Simulations are provided to demonstrate the efficiency of the proposed algorithms.

A fast kernel dimension reduction algorithm with applications to face recognition

This paper presents a novel dimensionality reduction algorithm for kernel based classification. In the feature space, the proposed algorithm maximizes the ratio of the squared between-class distance and the sum of the within-class variances of the training samples for a given reduced dimension. This algorithm has lower complexity than the recently reported kernel dimension reduction(KDR) for supervised learning. We conducted several simulations with large training datasets, which demonstrate that the proposed algorithm has similar performance or is marginally better compared with KDR whilst having the advantage of computational efficiency. Further, we applied the proposed dimension reduction algorithm to face recognition in which the number of training samples is very small. This proposed face recognition approach based on the new algorithm outperforms the eigenface approach based on the principle component analysis (PCA), when the training data is complete, that is, representative of the whole dataset.

Graph matching : fast candidate elimination using machine learning techniques

Graph matching is an important class of methods in pattern recognition. Typically, a graph representing an unknown pattern is matched with a database of models. If the database of model graphs is large, an additional factor in induced into the overall complexity of the matching process. Various techniques for reducing the influence of this additional factor have been described in the literature. In this paper we propose to extract simple features from a graph and use them to eliminate candidate graphs from the database. The most powerful set of features and a decision tree useful for candidate elimination are found by means of the C4.5 algorithm, which was originally proposed for inductive learning of classication rules. Experimental results are reported demonstrating that effcient candidate elimination can be achieved by the proposed procedure.

Influences on consumption of soft drinks and fast foods in adolescents

Soft drink and fast food are energy dense foodstuffs that are heavily marketed to adolescents, and are likely to be important in terms of risk of obesity. This study sought to examine the influences on soft drink and fast food consumption among adolescents as part of a cross-sectional survey of 2,719 adolescents (aged 11-16) from 93 randomly selected schools in New South Wales, Australia. Students provided information on soft drink and fast food consumption, and responded to statements examining influences over consumption. Over half of the boys and more than one third of the girls reported drinking soft drink daily, and consumption peaked in Grade 8 students. A quarter of students reported choosing soft drinks instead of water or milk, and around 40% agreed that soft drink was usually available in their homes. Availability in the home and drinking soft drinks with meals was most strongly associated with consumption in all age groups. Fast food consumption was higher among boys than girls in all age groups. Convenience and value for money yielded the strongest associations with fast food
consumption in boys, while preferring fast food to meals at home and preferring to “upsize” meals were most strongly associated with consumption in girls. Interventions to reduce consumption of soft drinks should target availability in both the home and school environment by removing soft drinks and replacing them with more nutritive beverages. Fast food outlets should be encouraged to provide a greater range of healthy and competitively priced options in reasonable portions.

Rapid depletion of genotypes with fast growth and bold personality traits from harvested fish populations

The possibility for fishery-induced evolution of life history traits is an important but unresolved issue for exploited fish populations. Because fisheries tend to select and remove the largest individuals, there is the evolutionary potential for lasting effects on fish production and productivity. Size selection represents an indirect mechanism of selection against rapid growth rate, because individual fish may be large because of rapid growth or because of slow growth but old age. The possibility for direct selection on growth rate, whereby fast-growing genotypes are more vulnerable to fishing irrespective of their size, is unexplored. In this scenario, faster-growing genotypes may be more vulnerable to fishing because of greater appetite and correspondingly greater feeding-related activity rates and boldness that could increase encounter with fishing gear and vulnerability to it. In a realistic whole-lake experiment, we show that fast-growing fish genotypes are harvested at three times the rate of the slow-growing genotypes within two replicate lake populations. Overall, 50% of fast-growing individuals were harvested compared with 30% of slow-growing individuals, independent of body size. Greater harvest of fast-growing genotypes was attributable to their greater behavioral vulnerability, being more active and bold. Given that growth is heritable in fishes, we speculate that evolution of slower growth rates attributable to behavioral vulnerability may be widespread in harvested fish populations. Our results indicate that commonly used minimum size-limits will not prevent overexploitation of fast-growing genotypes and individuals because of size-independent growth-rate selection by fishing.

Adaptive fast finite-time multiple-surface sliding control for a class of uncertain non-linear systems

This paper concerns the adaptive fast finite-time multiple-surface sliding control (AFFTMSSC) problem for a class of high-order uncertain non-linear systems of which the upper bounds of the system uncertainties are unknown. By using the fast control Lyapunov function and the method of so-called adding a power integrator merging with adaptive technique, a recursive design procedure is provided, which guarantees the fast finite-time stability of the closed-loop system. Further, it is proved that the control input is bounded.

Association between fast food purchasing and the local food environment

Objective: In this study, an instrument was created to measure the healthy and unhealthy characteristics of food environments and investigate associations between the whole of the food environment and fast food consumption.

Design and subjects: In consultation with other academic researchers in this field, food stores were categorised to either healthy or unhealthy and weighted (between +10 and −10) by their likely contribution to healthy/unhealthy eating practices. A healthy and unhealthy food environment score (FES) was created using these weightings. Using a cross-sectional study design, multilevel multinomial regression was used to estimate the effects of the whole food environment on the fast food purchasing habits of 2547 individuals.

Results: Respondents in areas with the highest tertile of the healthy FES had a lower likelihood of purchasing fast food both infrequently and frequently compared with respondents who never purchased, however only infrequent purchasing remained significant when simultaneously modelled with the unhealthy FES (odds ratio (OR) 0.52; 95% confidence interval (CI) 0.32–0.83). Although a lower likelihood of frequent fast food purchasing was also associated with living in the highest tertile of the unhealthy FES, no association remained once the healthy FES was included in the models. In our binary models, respondents living in areas with a higher unhealthy FES than healthy FES were more likely to purchase fast food infrequently (OR 1.35; 95% CI 1.00–1.82) however no association was found for frequent purchasing.

Conclusion: Our study provides some evidence to suggest that healthier food environments may discourage fast food purchasing.