To select or to weigh: a comparative study of linear combination schemes for superparent-one-dependence estimators

We conduct a large-scale comparative study on linearly combining superparent-one-dependence estimators (SPODEs), a popular family of seminaive Bayesian classifiers. Altogether, 16 model selection and weighing schemes, 58 benchmark data sets, and various statistical tests are employed. This paper's main contributions are threefold. First, it formally presents each scheme's definition, rationale, and time complexity and hence can serve as a comprehensive reference for researchers interested in ensemble learning. Second, it offers bias-variance analysis for each scheme's classification error performance. Third, it identifies effective schemes that meet various needs in practice. This leads to accurate and fast classification algorithms which have an immediate and significant impact on real-world applications. Another important feature of our study is using a variety of statistical tests to evaluate multiple learning methods across multiple data sets.

Analysis of linear networks with inconsistent initial conditions

This paper presents a new method to analyze timeinvariant linear networks allowing the existence of inconsistent initial conditions. This method is based on the use of distributions and state equations. Any time-invariant linear network can be analyzed. The network can involve any kind of pure or controlled sources. Also, the transferences of energy that occur at t=O are determined, and the concept of connection energy is introduced. The algorithms are easily implemented in a computer program.

Linear least squares estimation of the first order moving average parameter

We propose an iterative procedure to minimize the sum of squares function which avoids the nonlinear nature of estimating the first order moving average parameter and provides a closed form of the estimator. The asymptotic properties of the method are discussed and the consistency of the linear least squares estimator is proved for the invertible case. We perform various Monte Carlo experiments in order to compare the sample properties of the linear least squares estimator with its nonlinear counterpart for the conditional and unconditional cases. Some examples are also discussed

A theoretical and practical study on linear reforms of dual taxes

[cat] En aquest treball extenem les reformes lineals introduïdes per Pfähler (1984) al cas d’impostos duals. Estudiem l’efecte relatiu que els retalls lineals duals d’un impost dual tenen sobre la distribució de la desigualtat -es pot fer un estudi simètric per al cas d’augments d’impostos-. Tambe introduïm mesures del grau de progressivitat d’impostos duals i mostrem que estan connectades amb el criteri de dominació de Lorenz. Addicionalment, estudiem l’elasticitat de la càrrega fiscal de cadascuna de les reformes proposades. Finalment, gràcies a un model de microsimulació i una gran base de dades que conté informació sobre l’IRPF espanyol de l’any 2004, 1) comparem l’efecte que diferents reformes tindrien sobre l’impost dual espanyol i 2) estudiem quina redistribució de la riquesa va suposar la reforma dual de l’IRPF (Llei ’35/2006’) respecte l’anterior impost.

Respiratory chain dysfunction associated with multiple mitochondrial DNA deletions in antiretroviral therapy-related lipodystrophy.

Highly-active antiretroviral therapy (HAART) can induce a characteristic lipodystrophy syndrome characterized by peripheral fat wasting and central adiposity, usually associated with hyperlipidaemia and insulin resistance [1,2]. Indirect data have led some authors to propose that mitochondrial dysfunction could play a role in this syndrome [3,4].To date, as recently outlined by Kakuda et al. [5] in this journal, HIV-infected patients developing lipodystrophy have not been studied for mitochondrial changes or respiratory chain capacity...

Price transmission and market power analysis in the Spanish seafood market chain

[cat] Espanya és un dels principals mercats de productes pesquers d’Europa i del món. El consum de productes pesquers ha estat tradicionalment molt important a Espanya, el 2005 es varen consumir 36,7 kg per persona (MAPA, diversos anys). Malgrat això, el mercat i cóm interactuen els diversos nivells de la cadena de comercialització han gaudit de poca atenció. En aquest estudi, utilitzant dades setmanals, s’analitza per als dotze principals productes pesquers, l’elasticitat en la transmissió de preus al llarg de la cadena de comercialització a Espanya (llotja, mercat central i detallista). Finalment s’investiga la presència d’assimetria en la transmissió de preus entre aquests nivells de mercat. Els resultats obtinguts tenen importants implicacions a l’hora d’analitzar la demanda, poder de mercat i marges al llarg del mercat per als productes pesquers.

Electronic structure and magnetic interactions of the spin-chain compounds Ca2CuO3 and Sr2CuO3

The results are presented of a combined periodic and cluster model approach to the electronic structure and magnetic interactions in the spin-chain compounds Ca2CuO3 and Sr2CuO3. An extended t-J model is presented that includes in-chain and interchain hopping and magnetic interaction processes with parameters extracted from ab initio calculations. For both compounds, the in-chain magnetic interaction is found to be around -240 meV, larger than in any of the other cuprates reported in the literature. The interchain magnetic coupling is found to be weakly antiferromagnetic, -1 meV. The effective in-chain hopping parameters are estimated to be ~650 meV for both compounds, whereas the value of the interchain hopping parameter is 30 meV for Sr2CuO3 and 40 meV for Ca2CuO3, in line with the larger interchain distance in the former compound. These effective parameters are shown to be consistent with expressions recently suggested for the Néel temperature and the magnetic moments, and with relations that emerge from the t-J model Hamiltonian. Next, we investigate the physical nature of the band gap. Periodic calculations indicate that an interpretation in terms of a charge-transfer insulator is the most appropriate one, in contrast to the suggestion of a covalent correlated insulator recently reported in the literature.

Polydispersity index from linear viscoelastic data: Unimodal and bimodal linear polymer melts

This article describes a method for determining the polydispersity index Ip2=Mz/Mw of the molecular weight distribution (MWD) of linear polymeric materials from linear viscoelastic data. The method uses the Mellin transform of the relaxation modulus of a simple molecular rheological model. One of the main features of this technique is that it enables interesting MWD information to be obtained directly from dynamic shear experiments. It is not necessary to achieve the relaxation spectrum, so the ill-posed problem is avoided. Furthermore, a determinate shape of the continuous MWD does not have to be assumed in order to obtain the polydispersity index. The technique has been developed to deal with entangled linear polymers, whatever the form of the MWD is. The rheological information required to obtain the polydispersity index is the storage G′(ω) and loss G″(ω) moduli, extending from the terminal zone to the plateau region. The method provides a good agreement between the proposed theoretical approach and the experimental polydispersity indices of several linear polymers for a wide range of average molecular weights and polydispersity indices. It is also applicable to binary blends.

Combined effect of natural antimicrobials and high pressure processing to prevent Listeria monocytogenes growth after a cold chain break during storage of cooked ham

The effect of high pressure processing (400 MPa for 10 min) and natural antimicrobials 2 (enterocins and lactate-diacetate) on the behaviour of L. monocytogenes in sliced cooked ham 3 during refrigerated storage (1ºC and 6ºC) was assessed. The efficiency of the treatments after a 4 cold chain break was evaluated. Lactate-diacetate exerted a bacteriostatic effect against L. 5 monocytogenes during the whole storage period (3 months) at 1ºC and 6ºC, even after 6 temperature abuse. The combination of low storage temperature (1ºC), high pressure 7 processing (HPP) and addition of lactate-diacetate reduced the levels of L. monocytogenes 8 during storage by 2.7 log CFU/g. The most effective treatment was the combination of HPP, 9 enterocins and refrigeration at 1ºC, which reduced the population of the pathogen to final counts 10 of 4 MPN/g after 3 months of storage, even after the cold chain break.

On the tractability of the piecewise-linear approximation for general discrete-choice network revenue management

The choice network revenue management (RM) model incorporates customer purchase behavioras customers purchasing products with certain probabilities that are a function of the offeredassortment of products, and is the appropriate model for airline and hotel network revenuemanagement, dynamic sales of bundles, and dynamic assortment optimization. The underlyingstochastic dynamic program is intractable and even its certainty-equivalence approximation, inthe form of a linear program called Choice Deterministic Linear Program (CDLP) is difficultto solve in most cases. The separation problem for CDLP is NP-complete for MNL with justtwo segments when their consideration sets overlap; the affine approximation of the dynamicprogram is NP-complete for even a single-segment MNL. This is in contrast to the independentclass(perfect-segmentation) case where even the piecewise-linear approximation has been shownto be tractable. In this paper we investigate the piecewise-linear approximation for network RMunder a general discrete-choice model of demand. We show that the gap between the CDLP andthe piecewise-linear bounds is within a factor of at most 2. We then show that the piecewiselinearapproximation is polynomially-time solvable for a fixed consideration set size, bringing itinto the realm of tractability for small consideration sets; small consideration sets are a reasonablemodeling tradeoff in many practical applications. Our solution relies on showing that forany discrete-choice model the separation problem for the linear program of the piecewise-linearapproximation can be solved exactly by a Lagrangian relaxation. We give modeling extensionsand show by numerical experiments the improvements from using piecewise-linear approximationfunctions.

SAT Modulo Linear Arithmetic for Solving Polynomial

Polynomial constraint solving plays a prominent role in several areas of hardware and software analysis and verification, e.g., termination proving, program invariant generation and hybrid system verification, to name a few. In this paper we propose a new method for solving non-linear constraints based on encoding the problem into an SMT problem considering only linear arithmetic. Unlike other existing methods, our method focuses on proving satisfiability of the constraints rather than on proving unsatisfiability, which is more relevant in several applications as we illustrate with several examples. Nevertheless, we also present new techniques based on the analysis of unsatisfiable cores that allow one to efficiently prove unsatisfiability too for a broad class of problems. The power of our approach is demonstrated by means of extensive experiments comparing our prototype with state-of-the-art tools on benchmarks taken both from the academic and the industrial world.

Maximum likelihood Linear Programming Data Fusion for Speaker Recognition

Biometric system performance can be improved by means of data fusion. Several kinds of information can be fused in order to obtain a more accurate classification (identification or verification) of an input sample. In this paper we present a method for computing the weights in a weighted sum fusion for score combinations, by means of a likelihood model. The maximum likelihood estimation is set as a linear programming problem. The scores are derived from a GMM classifier working on a different feature extractor. Our experimental results assesed the robustness of the system in front a changes on time (different sessions) and robustness in front a change of microphone. The improvements obtained were significantly better (error bars of two standard deviations) than a uniform weighted sum or a uniform weighted product or the best single classifier. The proposed method scales computationaly with the number of scores to be fussioned as the simplex method for linear programming.

A non-linear VAD for noisy environments

This paper deals with non-linear transformations for improving the performance of an entropy-based voice activity detector (VAD). The idea to use a non-linear transformation has already been applied in the field of speech linear prediction, or linear predictive coding (LPC), based on source separation techniques, where a score function is added to classical equations in order to take into account the true distribution of the signal. We explore the possibility of estimating the entropy of frames after calculating its score function, instead of using original frames. We observe that if the signal is clean, the estimated entropy is essentially the same; if the signal is noisy, however, the frames transformed using the score function may give entropy that is different in voiced frames as compared to nonvoiced ones. Experimental evidence is given to show that this fact enables voice activity detection under high noise, where the simple entropy method fails.

Non-Linear and Non-Conventional Speech Processing: Alternative Techniques

This special issue aims to cover some problems related to non-linear and nonconventional speech processing. The origin of this volume is in the ISCA Tutorial and Research Workshop on Non-Linear Speech Processing, NOLISP’09, held at the Universitat de Vic (Catalonia, Spain) on June 25–27, 2009. The series of NOLISP workshops started in 2003 has become a biannual event whose aim is to discuss alternative techniques for speech processing that, in a sense, do not fit into mainstream approaches. A selected choice of papers based on the presentations delivered at NOLISP’09 has given rise to this issue of Cognitive Computation.