Design of neuro-fuzzy models by evolutionary and gradient-based algorithms

All systems found in nature exhibit, with different degrees, a nonlinear behavior. To emulate this behavior, classical systems identification techniques use, typically, linear models, for mathematical simplicity. Models inspired by biological principles (artificial neural networks) and linguistically motivated (fuzzy systems), due to their universal approximation property, are becoming alternatives to classical mathematical models. In systems identification, the design of this type of models is an iterative process, requiring, among other steps, the need to identify the model structure, as well as the estimation of the model parameters. This thesis addresses the applicability of gradient-basis algorithms for the parameter estimation phase, and the use of evolutionary algorithms for model structure selection, for the design of neuro-fuzzy systems, i.e., models that offer the transparency property found in fuzzy systems, but use, for their design, algorithms introduced in the context of neural networks. A new methodology, based on the minimization of the integral of the error, and exploiting the parameter separability property typically found in neuro-fuzzy systems, is proposed for parameter estimation. A recent evolutionary technique (bacterial algorithms), based on the natural phenomenon of microbial evolution, is combined with genetic programming, and the resulting algorithm, bacterial programming, advocated for structure determination. Different versions of this evolutionary technique are combined with gradient-based algorithms, solving problems found in fuzzy and neuro-fuzzy design, namely incorporation of a-priori knowledge, gradient algorithms initialization and model complexity reduction.

Biochemical indices and life traits of loggerhead turtles (Caretta caretta) from Cape Verde Islands

The loggerhead turtle (Caretta caretta) is an endangered marine reptile for whom assessing population health requires knowledge of demographic parameters such as individual growth rate. In Cape Verde, as within several populations, adult female loggerhead sea turtles show a size-related behavioral and trophic dichotomy. While smaller females are associated with oceanic habitats, larger females tend to feed in neritic habitats, which is reflected in their physiological condition and in their offspring. The ratio of RNA/DNA provides a measure of cellular protein synthesis capacity, which varies depending on changes in environmental conditions such as temperature and food availability. The purpose of this study was to evaluate the combined use of morphometric data and biochemical indices as predictors of the physiological condition of the females of distinct sizes and hatchlings during their nesting season and how temperature may influence the physiological condition on the offspring. Here we employed biochemical indices based on nucleic acid derived indices (standardized RNA/DNA ratio-sRD, RNA concentration and DNA concentration) in skin tissue as a potential predictor of recent growth rate in nesting females and hatchling loggerhead turtles. Our major findings were that the physiological condition of all nesting females (sRD) decreased during the nesting season, but that females associated with neritic habitats had a higher physiological condition than females associated with oceanic habitats. In addition, the amount of time required for a hatchling to right itself was negatively correlated with its physiological condition (sRD) and shaded nests produced hatchlings with lower sRD. Overall, our results showed that nucleic acid concentrations and ratios of RNA to DNA are an important tool as potential biomarkers of recent growth in marine turtles. Hence, as biochemical indices of instantaneous growth are likely temperature-, size- and age-dependent, the utility and validation of these indices on marine turtles stocks deserves further study.