Efficient feature selection filters for high-dimensional data

Feature selection is a central problem in machine learning and pattern recognition. On large datasets (in terms of dimension and/or number of instances), using search-based or wrapper techniques can be cornputationally prohibitive. Moreover, many filter methods based on relevance/redundancy assessment also take a prohibitively long time on high-dimensional. datasets. In this paper, we propose efficient unsupervised and supervised feature selection/ranking filters for high-dimensional datasets. These methods use low-complexity relevance and redundancy criteria, applicable to supervised, semi-supervised, and unsupervised learning, being able to act as pre-processors for computationally intensive methods to focus their attention on smaller subsets of promising features. The experimental results, with up to 10(5) features, show the time efficiency of our methods, with lower generalization error than state-of-the-art techniques, while being dramatically simpler and faster.

Materials selection for a set of multiple parts considering manufacturing costs and weight reduction with structural isoperformance using direct multisearch optimization

Materials selection is a matter of great importance to engineering design and software tools are valuable to inform decisions in the early stages of product development. However, when a set of alternative materials is available for the different parts a product is made of, the question of what optimal material mix to choose for a group of parts is not trivial. The engineer/designer therefore goes about this in a part-by-part procedure. Optimizing each part per se can lead to a global sub-optimal solution from the product point of view. An optimization procedure to deal with products with multiple parts, each with discrete design variables, and able to determine the optimal solution assuming different objectives is therefore needed. To solve this multiobjective optimization problem, a new routine based on Direct MultiSearch (DMS) algorithm is created. Results from the Pareto front can help the designer to align his/hers materials selection for a complete set of materials with product attribute objectives, depending on the relative importance of each objective.

Calibration, selection and mosaicing of SMART-1 AMIE images

Dissertação para obtenção do Grau de Mestre em Engenharia Electrotécnica e de Computadores

MOEA/VAN: Multiobjective Evolutionary Algorithm based on Vector Angle Neighborhood

Natural selection favors the survival and reproduction of organisms that are best adapted to their environment. Selection mechanism in evolutionary algorithms mimics this process, aiming to create environmental conditions in which artificial organisms could evolve solving the problem at hand. This paper proposes a new selection scheme for evolutionary multiobjective optimization. The similarity measure that defines the concept of the neighborhood is a key feature of the proposed selection. Contrary to commonly used approaches, usually defined on the basis of distances between either individuals or weight vectors, it is suggested to consider the similarity and neighborhood based on the angle between individuals in the objective space. The smaller the angle, the more similar individuals. This notion is exploited during the mating and environmental selections. The convergence is ensured by minimizing distances from individuals to a reference point, whereas the diversity is preserved by maximizing angles between neighboring individuals. Experimental results reveal a highly competitive performance and useful characteristics of the proposed selection. Its strong diversity preserving ability allows to produce a significantly better performance on some problems when compared with stat-of-the-art algorithms.

Predicting SO2 pollution incidents by means of additive models with optimum variable selection

The aim of this paper is to predict time series of SO2 concentrations emitted by coal-fired power stations in order to estimate in advance emission episodes and analyze the influence of some meteorological variables in the prediction. An emission episode is said to occur when the series of bi-hourly means of SO2 is greater than a specific level. For coal-fired power stations it is essential to predict emission epi- sodes sufficiently in advance so appropriate preventive measures can be taken. We proposed a meth- odology to predict SO2 emission episodes based on using an additive model and an algorithm for variable selection. The methodology was applied to the estimation of SO2 emissions registered in sampling lo- cations near a coal-fired power station located in Northern Spain. The results obtained indicate a good performance of the model considering only two terms of the time series and that the inclusion of the meteorological variables in the model is not significant.

Selection of patients with germline MLH1 mutated Lynch syndrome by determination of MLH1 methylation and BRAF mutation.

Lynch syndrome is one of the most common hereditary colorectal cancer (CRC) syndrome and is caused by germline mutations of MLH1, MSH2 and more rarely MSH6, PMS2, MLH3 genes. Whereas the absence of MSH2 protein is predictive of Lynch syndrome, it is not the case for the absence of MLH1 protein. The purpose of this study was to develop a sensitive and cost effective algorithm to select Lynch syndrome cases among patients with MLH1 immunohistochemical silencing. Eleven sporadic CRC and 16 Lynch syndrome cases with MLH1 protein abnormalities were selected. The BRAF c.1799T> A mutation (p.Val600Glu) was analyzed by direct sequencing after PCR amplification of exon 15. Methylation of MLH1 promoter was determined by Methylation-Sensitive Single-Strand Conformation Analysis. In patients with Lynch syndrome, there was no BRAF mutation and only one case showed MLH1 methylation (6%). In sporadic CRC, all cases were MLH1 methylated (100%) and 8 out of 11 cases carried the above BRAF mutation (73%) whereas only 3 cases were BRAF wild type (27%). We propose the following algorithm: (1) no further molecular analysis should be performed for CRC exhibiting MLH1 methylation and BRAF mutation, and these cases should be considered as sporadic CRC; (2) CRC with unmethylated MLH1 and negative for BRAF mutation should be considered as Lynch syndrome; and (3) only a small fraction of CRC with MLH1 promoter methylation but negative for BRAF mutation should be true Lynch syndrome patients. These potentially Lynch syndrome patients should be offered genetic counselling before searching for MLH1 gene mutations.

Clonotype selection and composition of human CD8 T cells specific for persistent herpes viruses varies with differentiation but is stable over time.

Protection from reactivation of persistent herpes virus infection is mediated by Ag-specific CD8 T cell responses, which are highly regulated by still poorly understood mechanisms. In this study, we analyzed differentiation and clonotypic dynamics of EBV- and CMV-specific T cells from healthy adults. Although these T lymphocytes included all subsets, from early-differentiated (EM/CD28(pos)) to late-differentiated (EMRA/CD28(neg)) stages, they varied in the sizes/proportions of these subsets. In-depth clonal composition analyses revealed TCR repertoires, which were highly restricted for CMV- and relatively diverse for EBV-specific cells. Virtually all virus-specific clonotypes identified in the EMRA/CD28(neg) subset were also found within the pool of less differentiated "memory" cells. However, striking differences in the patterns of dominance were observed among these subsets, because some clonotypes were selected with differentiation while others were not. Late-differentiated CMV-specific clonotypes were mostly characterized by TCR with lower dependency on CD8 coreceptor interaction. Yet all clonotypes displayed similar functional avidities, suggesting a compensatory role of CD8 in the clonotypes of lower TCR avidity. Importantly, clonotype selection and composition of each virus-specific subset upon differentiation was highly preserved over time, with the presence of the same dominant clonotypes at specific differentiation stages within a period of 4 years. Remarkably, clonotypic distribution was stable not only in late-differentiated but also in less-differentiated T cell subsets. Thus, T cell clonotypes segregate with differentiation, but the clonal composition once established is kept constant for at least several years. These findings reveal novel features of the highly sophisticated control of steady state protective T cell activity in healthy adults.

Time Series Input Selection using Multiple Kernel Learning

In this paper we study the relevance of multiple kernel learning (MKL) for the automatic selection of time series inputs. Recently, MKL has gained great attention in the machine learning community due to its flexibility in modelling complex patterns and performing feature selection. In general, MKL constructs the kernel as a weighted linear combination of basis kernels, exploiting different sources of information. An efficient algorithm wrapping a Support Vector Regression model for optimizing the MKL weights, named SimpleMKL, is used for the analysis. In this sense, MKL performs feature selection by discarding inputs/kernels with low or null weights. The approach proposed is tested with simulated linear and nonlinear time series (AutoRegressive, Henon and Lorenz series).

Impact of recombination on genetic variability within Staphylococcus aureus clonal complexes.

The population structure of Staphylococcus aureus is generally described as highly clonal and is consequently subdivided into several clonal complexes (CCs). Recent data suggested that recombination might occur more frequently within than among CCs. To test this hypothesis as well as to understand how genetic diversity is created in S. aureus, we analyzed a collection of 182 isolates with MLST and five highly variable core adhesion (ADH) genes. As expected the polymorphism of ADH genes was higher than MLST genes. However both categories of genes showed low within CCs diversity with a dominant haplotype and its single nucleotide variants. Several recombination events were detected but none involved intra-CC recombination. This did not confirm the hypothesis of higher recombination within CCs. Nevertheless, molecular analyses of variance indicated that these few recombination events have a significant impact on the genetic diversity within CCs. In addition, although most ADH genes were under purifying selection, signs of positive selection associated with a recombinant group were detected. These data highlight the importance of recombination on the evolution of the highly clonal S. aureus and suggest that recombination when combined with demographic mechanisms as well as selection might favor the rapid creation of new clonal complexes.

TCR-alpha CDR3 loop audition regulates positive selection.

How positive selection molds the T cell repertoire has been difficult to examine. In this study, we use TCR-beta-transgenic mice in which MHC shapes TCR-alpha use. Differential AV segment use is directly related to the constraints placed on the composition of the CDR3 loops. Where these constraints are low, efficient selection of alphabeta pairs follows. This mode of selection preferentially uses favored AV-AJ rearrangements and promotes diversity. Increased constraint on the alpha CDR3 loops leads to inefficient selection associated with uncommon recombination events and limited diversity. Further, the two modes of selection favor alternate sets of AJ segments. We discuss the relevance of these findings to the imprint of self-MHC restriction and peripheral T cell activation.

Quantitative impact of thymic clonal deletion on the T cell repertoire.

Interactions between major histocompatibility complex (MHC) molecules expressed on stromal cells and antigen-specific receptors on T cells shape the repertoire of mature T lymphocytes emerging from the thymus. Some thymocytes with appropriate receptors are stimulated to undergo differentiation to the fully mature state (positive selection), whereas others with strongly autoreactive receptors are triggered to undergo programmed cell death before completing this differentiation process (negative selection). The quantitative impact of negative selection on the potentially available repertoire is currently unknown. To address this issue, we have constructed radiation bone marrow chimeras in which MHC molecules are present on radioresistant thymic epithelial cells (to allow positive selection) but absent from radiosensitive hematopoietic elements responsible for negative selection. In such chimeras, the number of mature thymocytes was increased by twofold as compared with appropriate control chimeras This increase in steady-state numbers of mature thymocytes was not related to proliferation, increased retention, or recirculation and was accompanied by a similar two- to threefold increase in the de novo rate of generation of mature cells. Taken together, our data indicate that half to two-thirds of the thymocytes able to undergo positive selection die before full maturation due to negative selection.

Promoter IV of the class II transactivator gene is essential for positive selection of CD4+ T cells.

Major histocompatibility complex class II (MHCII) expression is regulated by the transcriptional coactivator CIITA. Positive selection of CD4(+) T cells is abrogated in mice lacking one of the promoters (pIV) of the Mhc2ta gene. This is entirely due to the absence of MHCII expression in thymic epithelia, as demonstrated by bone marrow transfer experiments between wild-type and pIV(-/-) mice. Medullary thymic epithelial cells (mTECs) are also MHCII(-) in pIV(-/-) mice. Bone marrow-derived, professional antigen-presenting cells (APCs) retain normal MHCII expression in pIV(-/-) mice, including those believed to mediate negative selection in the thymic medulla. Endogenous retroviruses thus retain their ability to sustain negative selection of the residual CD4(+) thymocytes in pIV(-/-) mice. Interestingly, the passive acquisition of MHCII molecules by thymocytes is abrogated in pIV(-/-) mice. This identifies thymic epithelial cells as the source of this passive transfer. In peripheral lymphoid organs, the CD4(+) T-cell population of pIV(-/-) mice is quantitatively and qualitatively comparable to that of MHCII-deficient mice. It comprises a high proportion of CD1-restricted natural killer T cells, which results in a bias of the V beta repertoire of the residual CD4(+) T-cell population. We have also addressed the identity of the signal that sustains pIV expression in cortical epithelia. We found that the Jak/STAT pathways activated by the common gamma chain (CD132) or common beta chain (CDw131) cytokine receptors are not required for MHCII expression in thymic cortical epithelia.

Three essays in financial economics: asset pricing, optimal portfolio selection and financial integration

Executive Summary The unifying theme of this thesis is the pursuit of a satisfactory ways to quantify the riskureward trade-off in financial economics. First in the context of a general asset pricing model, then across models and finally across country borders. The guiding principle in that pursuit was to seek innovative solutions by combining ideas from different fields in economics and broad scientific research. For example, in the first part of this thesis we sought a fruitful application of strong existence results in utility theory to topics in asset pricing. In the second part we implement an idea from the field of fuzzy set theory to the optimal portfolio selection problem, while the third part of this thesis is to the best of our knowledge, the first empirical application of some general results in asset pricing in incomplete markets to the important topic of measurement of financial integration. While the first two parts of this thesis effectively combine well-known ways to quantify the risk-reward trade-offs the third one can be viewed as an empirical verification of the usefulness of the so-called "good deal bounds" theory in designing risk-sensitive pricing bounds. Chapter 1 develops a discrete-time asset pricing model, based on a novel ordinally equivalent representation of recursive utility. To the best of our knowledge, we are the first to use a member of a novel class of recursive utility generators to construct a representative agent model to address some long-lasting issues in asset pricing. Applying strong representation results allows us to show that the model features countercyclical risk premia, for both consumption and financial risk, together with low and procyclical risk free rate. As the recursive utility used nests as a special case the well-known time-state separable utility, all results nest the corresponding ones from the standard model and thus shed light on its well-known shortcomings. The empirical investigation to support these theoretical results, however, showed that as long as one resorts to econometric methods based on approximating conditional moments with unconditional ones, it is not possible to distinguish the model we propose from the standard one. Chapter 2 is a join work with Sergei Sontchik. There we provide theoretical and empirical motivation for aggregation of performance measures. The main idea is that as it makes sense to apply several performance measures ex-post, it also makes sense to base optimal portfolio selection on ex-ante maximization of as many possible performance measures as desired. We thus offer a concrete algorithm for optimal portfolio selection via ex-ante optimization over different horizons of several risk-return trade-offs simultaneously. An empirical application of that algorithm, using seven popular performance measures, suggests that realized returns feature better distributional characteristics relative to those of realized returns from portfolio strategies optimal with respect to single performance measures. When comparing the distributions of realized returns we used two partial risk-reward orderings first and second order stochastic dominance. We first used the Kolmogorov Smirnov test to determine if the two distributions are indeed different, which combined with a visual inspection allowed us to demonstrate that the way we propose to aggregate performance measures leads to portfolio realized returns that first order stochastically dominate the ones that result from optimization only with respect to, for example, Treynor ratio and Jensen's alpha. We checked for second order stochastic dominance via point wise comparison of the so-called absolute Lorenz curve, or the sequence of expected shortfalls for a range of quantiles. As soon as the plot of the absolute Lorenz curve for the aggregated performance measures was above the one corresponding to each individual measure, we were tempted to conclude that the algorithm we propose leads to portfolio returns distribution that second order stochastically dominates virtually all performance measures considered. Chapter 3 proposes a measure of financial integration, based on recent advances in asset pricing in incomplete markets. Given a base market (a set of traded assets) and an index of another market, we propose to measure financial integration through time by the size of the spread between the pricing bounds of the market index, relative to the base market. The bigger the spread around country index A, viewed from market B, the less integrated markets A and B are. We investigate the presence of structural breaks in the size of the spread for EMU member country indices before and after the introduction of the Euro. We find evidence that both the level and the volatility of our financial integration measure increased after the introduction of the Euro. That counterintuitive result suggests the presence of an inherent weakness in the attempt to measure financial integration independently of economic fundamentals. Nevertheless, the results about the bounds on the risk free rate appear plausible from the view point of existing economic theory about the impact of integration on interest rates.

In vivo T-lymphocyte tolerance in the absence of thymic clonal deletion mediated by hematopoietic cells.

Thymic negative selection renders the developing T-cell repertoire tolerant to self-major histocompatability complex (MHC)/peptide ligands. The major mechanism of induction of self-tolerance is thought to be thymic clonal deletion, ie, the induction of apoptotic cell death in thymocytes expressing a self-reactive T-cell receptor. Consistent with this hypothesis, in mice deficient in thymic clonal deletion mediated by cells of hematopoietic origin, a twofold to threefold increased generation of mature thymocytes has been observed. Here we describe the analysis of the specificity of T lymphocytes developing in the absence of clonal deletion mediated by hematopoietic cells. In vitro, targets expressing syngeneic MHC were readily lysed by activated CD8(+) T cells from deletion-deficient mice. However, proliferative responses of T cells from these mice on activation with syngeneic antigen presenting cells were rather poor. In vivo, deletion-deficient T cells were incapable of induction of lethal graft-versus-host disease in syngeneic hosts. These data indicate that in the absence of thymic deletion mediated by hematopoietic cells functional T-cell tolerance can be induced by nonhematopoietic cells in the thymus. Moreover, our results emphasize the redundancy in thymic negative selection mechanisms.

Viral superantigen-induced negative selection of TCR transgenic CD4+ CD8+ thymocytes depends on activation, but not proliferation.

T-cell negative selection, a process by which intrathymic immunological tolerance is induced, involves the apoptosis-mediated clonal deletion of potentially autoreactive T cells. Although different experimental approaches suggest that this process is triggered as the result of activation-mediated cell death, the signal transduction pathways underlying this process is not fully understood. In the present report we have used an in vitro system to analyze the cell activation and proliferation requirements for the deletion of viral superantigen (SAg)-reactive Vbeta8.1 T-cell receptor (TCR) transgenic (TG) thymocytes. Our results indicate that in vitro negative selection of viral SAg-reactive CD4+ CD8+ thymocytes is dependent on thymocyte activation but does not require the proliferation of the negatively signaled thymocytes.