Genetic programming and bacterial algorithm for neural networks and fuzzy systems design

In the field of control systems it is common to use techniques based on model adaptation to carry out control for plants for which mathematical analysis may be intricate. Increasing interest in biologically inspired learning algorithms for control techniques such as Artificial Neural Networks and Fuzzy Systems is in progress. In this line, this paper gives a perspective on the quality of results given by two different biologically connected learning algorithms for the design of B-spline neural networks (BNN) and fuzzy systems (FS). One approach used is the Genetic Programming (GP) for BNN design and the other is the Bacterial Evolutionary Algorithm (BEA) applied for fuzzy rule extraction. Also, the facility to incorporate a multi-objective approach to the GP algorithm is outlined, enabling the designer to obtain models more adequate for their intended use.

B-spline and neuro-fuzzy models design with function and derivative equalities

The design of neuro-fuzzy models is still a complex problem, as it involves not only the determination of the model parameters, but also its structure. Of special importance is the incorporation of a priori information in the design process. In this paper two known design algorithms for B-spline models will be updated to account for function and derivatives equality restrictions, which are important when the neural model is used for performing single or multi-objective optimization on-line.

Training neuro-fuzzy models using evolution based algorithms

The normal design process for neural networks or fuzzy systems involve two different phases: the determination of the best topology, which can be seen as a system identification problem, and the determination of its parameters, which can be envisaged as a parameter estimation problem. This latter issue, the determination of the model parameters (linear weights and interior knots) is the simplest task and is usually solved using gradient or hybrid schemes. The former issue, the topology determination, is an extremely complex task, especially if dealing with real-world problems.

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.

Marine foundation species shifting ranges and genetic baselines

This thesis revealed the most importance factors shaping the distribution, abundance and genetic diversity of four marine foundation species. Environmental conditions, particularly sea temperatures, nutrient availability and ocean waves, played a primary role in shaping the spatial distribution and abundance of populations, acting on scales varying from tens of meters to hundreds of kilometres. Furthermore, the use of Species Distribution Models (SDMs) with biological records of occurrence and high-resolution oceanographic data, allowed predicting species distributions across time. This approach highlighted the role of climate change, particularly when extreme temperatures prevailed during glacial and interglacial periods. These results, when combined with mtDNA and microsatellite genetic variation of populations allowed inferring for the influence of past range dynamics in the genetic diversity and structure of populations. For instance, the Last Glacial Maximum produced important shifts in species ranges, leaving obvious signatures of higher genetic diversities in regions where populations persisted (i.e., refugia). However, it was found that a species’ genetic pool is shaped by regions of persistence, adjacent to others experiencing expansions and contractions. Contradicting expectations, refugia seem to play a minor role on the re(colonization) process of previously eroded populations. In addition, the available habitat area for expanding populations and the inherent mechanisms of species dispersal in occupying available habitats were also found to be fundamental in shaping the distributions of genetic diversity. However, results suggest that the high levels of genetic diversity in some populations do not rule out that they may have experienced strong genetic erosion in the past, a process here named shifting genetic baselines. Furthermore, this thesis predicted an ongoing retraction at the rear edges and extinctions of unique genetic lineages, which will impoverish the global gene pool, strongly shifting the genetic baselines in the future.

Environmental variables, habitat discontinuity and life history shaping the genetic structure of Pomatoschistus marmoratus

Coastal lagoons are semi-isolated ecosystems exposed to wide fluctuations of environmental conditions and showing habitat fragmentation. These features may play an important role in separating species into different populations, even at small spatial scales. In this study, we evaluate the concordance between mitochondrial (previous published data) and nuclear data analyzing the genetic variability of Pomatoschistus marmoratus in five localities, inside and outside the Mar Menor coastal lagoon (SE Spain) using eight microsatellites. High genetic diversity and similar levels of allele richness were observed across all loci and localities, although significant genic and genotypic differentiation was found between populations inside and outside the lagoon. In contrast to the FST values obtained from previous mitochondrial DNA analyses (control region), the microsatellite data exhibited significant differentiation among samples inside the Mar Menor and between lagoonal and marine samples. This pattern was corroborated using Cavalli-Sforza genetic distances. The habitat fragmentation inside the coastal lagoon and among lagoon and marine localities could be acting as a barrier to gene flow and contributing to the observed genetic structure. Our results from generalized additive models point a significant link between extreme lagoonal environmental conditions (mainly maximum salinity) and P. marmoratus genetic composition. Thereby, these environmental features could be also acting on genetic structure of coastal lagoon populations of P. marmoratus favoring their genetic divergence. The mating strategy of P. marmoratus could be also influencing our results obtained from mitochondrial and nuclear DNA. Therefore, a special consideration must be done in the selection of the DNA markers depending on the reproductive strategy of the species.

Genetic diversity and connectivity remain high in Holothuria polii (Delle Chiaje 1823) across a coastal lagoon-open sea environmental gradient

Coastal lagoons represent habitats with widely heterogeneous environmental conditions, particularly as regards salinity and temperature,which fluctuate in both space and time. These characteristics suggest that physical and ecological factors could contribute to the genetic divergence among populations occurring in coastal lagoon and opencoast environments. This study investigates the genetic structure of Holothuria polii at a micro-geographic scale across theMar Menor coastal lagoon and nearbymarine areas, estimating the mitochondrial DNA variation in two gene fragments, cytochrome oxidase I (COI) and 16S rRNA (16S). Dataset of mitochondrial sequences was also used to test the influence of environmental differences between coastal lagoon andmarine waters on population genetic structure. All sampled locations exhibited high levels of haplotype diversity and low values of nucleotide diversity. Both genes showed contrasting signals of genetic differentiation (non-significant differences using COI and slight differences using 16S, which could due to different mutation rates or to differential number of exclusive haplotypes. We detected an excess of recent mutations and exclusive haplotypes, which can be generated as a result of population growth. However, selective processes can be also acting on the gene markers used; highly significant generalized additive models have been obtained considering genetic data from16S gene and independent variables such as temperature and salinity.