Information flow analysis using data-dependent logical propositions

Dissertação para obtenção do Grau de Mestre em Engenharia Informática

A data-dependent skeleton estimate and a scale-sensitive dimension for classification

The classical binary classification problem is investigatedwhen it is known in advance that the posterior probability function(or regression function) belongs to some class of functions. We introduceand analyze a method which effectively exploits this knowledge. The methodis based on minimizing the empirical risk over a carefully selected``skeleton'' of the class of regression functions. The skeleton is acovering of the class based on a data--dependent metric, especiallyfitted for classification. A new scale--sensitive dimension isintroduced which is more useful for the studied classification problemthan other, previously defined, dimension measures. This fact isdemonstrated by performance bounds for the skeleton estimate in termsof the new dimension.

Experimental demonstration of data-dependent pilot-aided phase noise estimation for CO-OFDM

We demonstrate a novel phase noise estimation scheme for CO-OFDM, in which pilot subcarriers are deliberately correlated to the data subcarriers. This technique reduces the overhead by a factor of 2. © OSA 2014.

Predictors of nonresponse in a questionnaire-based outcome study of vocational rehabilitation patients.

OBJECTIVE: To identify predictors of nonresponse to a self-report study of patients with orthopedic trauma hospitalized for vocational rehabilitation between November 15, 2003, and December 31, 2005. The role of biopsychosocial complexity, assessed using the INTERMED, was of particular interest. DESIGN: Cohort study. Questionnaires with quality of life, sociodemographic, and job-related questions were given to patients at hospitalization and 1 year after discharge. Sociodemographic data, biopsychosocial complexity, and presence of comorbidity were available at hospitalization (baseline) for all eligible patients. Logistic regression models were used to test a number of baseline variables as potential predictors of nonresponse to the questionnaires at each of the 2 time points. SETTING: Rehabilitation clinic. PARTICIPANTS: Patients (N=990) hospitalized for vocational rehabilitation over a period of 2 years. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURE: Nonresponse to the questionnaires was the binary dependent variable. RESULTS: Patients with high biopsychosocial complexity, foreign native language, or low educational level were less likely to respond at both time points. Younger patients were less likely to respond at 1 year. Those living in a stable partnership were less likely than singles to respond at hospitalization. Sex, psychiatric, and somatic comorbidity and alcoholism were never associated with nonresponse. CONCLUSIONS: We stress the importance of assessing biopsychosocial complexity to predict nonresponse. Furthermore, the factors we found to be predictive of nonresponse are also known to influence treatment outcome and vocational rehabilitation. Therefore, it is important to increase the response rate of the groups of concern in order to reduce selection bias in epidemiologic investigations.

Subsampling the mean of heavy-tailed dependent observations

We establish the validity of subsampling confidence intervals for themean of a dependent series with heavy-tailed marginal distributions.Using point process theory, we study both linear and nonlinear GARCH-liketime series models. We propose a data-dependent method for the optimalblock size selection and investigate its performance by means of asimulation study.

Reusing values in a dynamic conditional execution architecture

A Execução Condicional Dinâmica (DCE) é uma alternativa para redução dos custos relacionados a desvios previstos incorretamente. A idéia básica é buscar todos os fluxos produzidos por um desvio que obedecem algumas restrições relativas à complexidade e tamanho. Como conseqüência, um número menor de previsões é executado, e assim, um número mais baixo de desvios é incorretamente previsto. Contudo, tal como outras soluções multi-fluxo, o DCE requer uma estrutura de controle mais complexa. Na arquitetura DCE, é observado que várias réplicas da mesma instrução são despachadas para as unidades funcionais, bloqueando recursos que poderiam ser utilizados por outras instruções. Essas réplicas são geradas após o ponto de convergência dos diversos fluxos em execução e são necessárias para garantir a semântica correta entre instruções dependentes de dados. Além disso, o DCE continua produzindo réplicas até que o desvio que gerou os fluxos seja resolvido. Assim, uma seção completa do código pode ser replicado, reduzindo o desempenho. Uma alternativa natural para esse problema é reusar essas seções (ou traços) que são replicadas. O objetivo desse trabalho é analisar e avaliar a efetividade do reuso de valores na arquitetura DCE. Como será apresentado, o princípio do reuso, em diferentes granularidades, pode reduzir efetivamente o problema das réplicas e levar a aumentos de desempenho.

Algorithms for Data Flows

* The research was supported by INTAS 00-397 and 00-626 Projects.

A World without Causation: Big Data and the Coming of Age of Posthumanism

Advocates of Big Data assert that we are in the midst of an epistemological revolution, promising the displacement of the modernist methodological hegemony of causal analysis and theory generation. It is alleged that the growing ‘deluge’ of digitally generated data, and the development of computational algorithms to analyse them, has enabled new inductive ways of accessing everyday relational interactions through their ‘datafication’. This paper critically engages with these discourses of Big Data and complexity, particularly as they operate in the discipline of International Relations, where it is alleged that Big Data approaches have the potential for developing self-governing societal capacities for resilience and adaptation through the real-time reflexive awareness and management of risks and problems as they arise. The epistemological and ontological assumptions underpinning Big Data are then analysed to suggest that critical and posthumanist approaches have come of age through these discourses, enabling process-based and relational understandings to be translated into policy and governance practices. The paper thus raises some questions for the development of critical approaches to new posthuman forms of governance and knowledge production.

Design of clinical pharmacology trials

1. There are a variety of methods that could be used to increase the efficiency of the design of experiments. However, it is only recently that such methods have been considered in the design of clinical pharmacology trials. 2. Two such methods, termed data-dependent (e.g. simulation) and data-independent (e.g. analytical evaluation of the information in a particular design), are becoming increasingly used as efficient methods for designing clinical trials. These two design methods have tended to be viewed as competitive, although a complementary role in design is proposed here. 3. The impetus for the use of these two methods has been the need for a more fully integrated approach to the drug development process that specifically allows for sequential development (i.e. where the results of early phase studies influence later-phase studies). 4. The present article briefly presents the background and theory that underpins both the data-dependent and -independent methods with the use of illustrative examples from the literature. In addition, the potential advantages and disadvantages of each method are discussed.

Building adaptable, QoS-aware dependable embedded systems

Most of today’s embedded systems are required to work in dynamic environments, where the characteristics of the computational load cannot always be predicted in advance. Furthermore, resource needs are usually data dependent and vary over time. Resource constrained devices may need to cooperate with neighbour nodes in order to fulfil those requirements and handle stringent non-functional constraints. This paper describes a framework that facilitates the distribution of resource intensive services across a community of nodes, forming temporary coalitions for a cooperative QoSaware execution. The increasing need to tailor provided service to each application’s specific needs determines the dynamic selection of peers to form such a coalition. The system is able to react to load variations, degrading its performance in a controlled fashion if needed. Isolation between different services is achieved by guaranteeing a minimal service quality to accepted services and by an efficient overload control that considers the challenges and opportunities of dynamic distributed embedded systems.

Improved nonparametric confidence intervals in time series regressions

Condence intervals in econometric time series regressions suffer fromnotorious coverage problems. This is especially true when the dependencein the data is noticeable and sample sizes are small to moderate, as isoften the case in empirical studies. This paper suggests using thestudentized block bootstrap and discusses practical issues, such as thechoice of the block size. A particular data-dependent method is proposedto automate the method. As a side note, it is pointed out that symmetricconfidence intervals are preferred over equal-tailed ones, since theyexhibit improved coverage accuracy. The improvements in small sampleperformance are supported by a simulation study.

Model selection and error estimation

We study model selection strategies based on penalized empirical loss minimization. We point out a tight relationship between error estimation and data-based complexity penalization: any good error estimate may be converted into a data-based penalty function and the performance of the estimate is governed by the quality of the error estimate. We consider several penalty functions, involving error estimates on independent test data, empirical {\sc vc} dimension, empirical {\sc vc} entropy, andmargin-based quantities. We also consider the maximal difference between the error on the first half of the training data and the second half, and the expected maximal discrepancy, a closely related capacity estimate that can be calculated by Monte Carlo integration. Maximal discrepancy penalty functions are appealing for pattern classification problems, since their computation is equivalent to empirical risk minimization over the training data with some labels flipped.

General unknown screening procedure for the characterization of human drug metabolites: application to loratadine phase I metabolism.

This article describes the application of a recently developed general unknown screening (GUS) strategy based on LC coupled to a hybrid linear IT-triple quadrupole mass spectrometer (LC-MS/MS-LIT) for the simultaneous detection and identification of drug metabolites following in vitro incubation with human liver microsomes. The histamine H1 receptor antagonist loratadine was chosen as a model compound to demonstrate the interest of such approach, because of its previously described complex and extensive metabolism. Detection and mass spectral characterization were based on data-dependent acquisition, switching between a survey scan acquired in the ion-trapping Q3 scan mode with dynamic subtraction of background noise, and a dependent scan in the ion-trapping product ion scan mode of automatically selected parent ions. In addition, the MS(3) mode was used in a second step to confirm the structure of a few fragment ions. The sensitivity of the ion-trapping modes combined with the selectivity of the triple quadrupole modes allowed, with only one injection, the detection and identification of 17 phase I metabolites of loratadine. The GUS procedure used in this study may be applicable as a generic technique for the characterization of drug metabolites after in vitro incubation, as well as probably in vivo experiments.

Fast Geodesic Active Fields for Image Registration Based on Splitting and Augmented Lagrangian Approaches.

In this paper, we present an efficient numerical scheme for the recently introduced geodesic active fields (GAF) framework for geometric image registration. This framework considers the registration task as a weighted minimal surface problem. Hence, the data-term and the regularization-term are combined through multiplication in a single, parametrization invariant and geometric cost functional. The multiplicative coupling provides an intrinsic, spatially varying and data-dependent tuning of the regularization strength, and the parametrization invariance allows working with images of nonflat geometry, generally defined on any smoothly parametrizable manifold. The resulting energy-minimizing flow, however, has poor numerical properties. Here, we provide an efficient numerical scheme that uses a splitting approach; data and regularity terms are optimized over two distinct deformation fields that are constrained to be equal via an augmented Lagrangian approach. Our approach is more flexible than standard Gaussian regularization, since one can interpolate freely between isotropic Gaussian and anisotropic TV-like smoothing. In this paper, we compare the geodesic active fields method with the popular Demons method and three more recent state-of-the-art algorithms: NL-optical flow, MRF image registration, and landmark-enhanced large displacement optical flow. Thus, we can show the advantages of the proposed FastGAF method. It compares favorably against Demons, both in terms of registration speed and quality. Over the range of example applications, it also consistently produces results not far from more dedicated state-of-the-art methods, illustrating the flexibility of the proposed framework.