A simplified accurate algorithm for the fisher-yates exact test

An algorithm, using a short cut due to Feldman and Klingler, for the Fisher-Yates exact test is presented: the algorithm is quick, simple and accurate.I am grateful to the Editor and two reviewers for a number of constructive and helpful suggestions.     

Piecewise method of reciprocal averages for dual scaling of multiple-choice data

The proposed method handles the classical method of reciprocal averages (MRA) in a piecewise (item-by-item) mode, whereby one can deal with smaller matrices and attain faster convergence to a solution than the MRA. A new concept the principle of constant proportionality is introduced to provide an interesting interpretation for scaling multiple-choice data a la Guttman. A small example is presented for discussion of the technique.This study was supported by the Natural Sciences and Engineering Research Council Canada Grant (No. A4581) to S. Nishisato. The authors are indebted to reviewers for valuable comments.     

Adaptive sampling for active learning with genetic programming

Active learning is a machine learning paradigm allowing to decide which inputs to use for training. It is introduced to Genetic Programming (GP) essentially thanks to the dynamic data sampling, used to address some known issues such as the computational cost, the over-fitting problem and the imbalanced databases. The traditional dynamic sampling for GP gives to the algorithm a new sample periodically, often each generation, without considering the state of the evolution. In so doing, individuals do not have enough time to extract the hidden knowledge. An alternative approach is to use some information about the learning state to adapt the periodicity of the training data change. In this work, we propose an adaptive sampling strategy for classification tasks based on the state of solved fitness cases throughout learning. It is a flexible approach that could be applied with any dynamic sampling. We implemented some sampling algorithms extended with dynamic and adaptive controlling re-sampling frequency. We experimented them to solve the KDD intrusion detection and the Adult incomes prediction problems with GP. The experimental study demonstrates how the sampling frequency control preserves the power of dynamic sampling with possible improvements in learning time and quality. We also demonstrate that adaptive sampling can be an alternative to multi-level sampling. This work opens many new relevant extension paths.     

Le malade génétique responsable « La loi des suspects »

A patient suffers of a genetical disease. He does not wish to inform his family. Is he responsible for a lost of chances of his parents, chances to prevent the disease, or to receive care? Is there a second class of patients?     

Bayesian Modal Estimation for the One-Parameter Logistic Ability-Based Guessing (1PL-AG) Model

The calibration of the one-parameter logistic ability-based guessing (1PL-AG) model in item response theory (IRT) with a modest sample size remains a challenge for its implausible estimates and difficulty in obtaining standard errors of estimates. This article proposes an alternative Bayesian modal estimation (BME) method, the Bayesian Expectation-Maximization-Maximization (BEMM) method, which is developed by combining an augmented variable formulation of the 1PL-AG model and a mixture model conceptualization of the three-parameter logistic model (3PLM). By comparing with marginal maximum likelihood estimation (MMLE) and Markov Chain Monte Carlo (MCMC) in JAGS, the simulation shows that BEMM can produce stable and accurate estimates in the modest sample size. A real data example and the MATLAB codes of BEMM are also provided.     

Local Optima in Mixture Modeling

It is common knowledge that mixture models are prone to arrive at locally optimal solutions. Typically, researchers are directed to utilize several random initializations to ensure that the resulting solution is adequate. However, it is unknown what factors contribute to a large number of local optima and whether these coincide with the factors that reduce the accuracy of a mixture model. A real-data illustration and a series of simulations are presented that examine the effect of a variety of data structures on the propensity of local optima and the classification quality of the resulting solution. We show that there is a moderately strong relationship between a solution that has a high proportion of local optima and one that is poorly classified.     

Cost minimization during simulated evolution of paired neural networks leads to asymmetries and specialization

Motivated by specialization (lateralization) that occurs in corresponding left and right regions of the cerebral cortex, several past computational models have studied conditions under which functional specialization can arise during learning due to underlying asymmetries in paired neural networks. However, these past studies have not addressed the basic issue of how such underlying asymmetries arise in the first place. As an initial step in addressing this issue, we investigated the hypothesis that underlying asymmetries will appear in paired neural networks during a simulated evolutionary process when fitness is based not only on maximizing performance, but also on minimizing various ‘costs’ such as energy consumption, neural connection weights, and response times. Simulated evolution under these conditions consistently produced networks with left–right asymmetries in region size, excitability and plasticity. These underlying asymmetries were often synergistic, leading to subsequent functional lateralization during network training. While our computational models are too simple for these results to be directly extrapolated to real nervous systems, they provide support for the hypothesis that brain asymmetries and lateralization in biological nervous systems may be a consequence of cost minimization present during evolution, and are the first computational demonstration of emergent population lateralization.     

A contribution to the development of a psychoanalytical methodology for research into language: A systematic study of narration as an expression of sexuality

Systematic research on language has a solid starting point in Freudian hypotheses on sexuality. This theory is the semantic ground for categorising narrations and providing a basis for the research method. The author places his research in the frame of the systematisation of the Freudian theory of substitute formations in the preconscious. He affirms that these formations are influenced in a particular way by each variant of sexuality. The author proposes an inventory (in Freudian terms) of those variants of sexuality, and he affirms that they can be detected in the discourse. Five universal scenes having the status of a canon, with specific features for each of the ways in which sexuality is manifested, make up narrations. The author describes the features of each narration and provides some examples. He also examines problems related to the use of the method (the coexistence of different variants of sexuality in a single fragment of a narration and the successive steps in the use of the method). The author briefly considers some other problems: the analysis of defences, the study of words and theoretical research. Finally, he examines the validity and reliability of the method.     

Performance based algorithms for video bit transmissions

The usage of video content has increased in past ten decades. As a result, increase in usage of commercial video coding standards called High Efficiency Video Coding (HEVC/H.265). A self-Adaptive N-depth Context Tree Weighting algorithm (SANDCTW) is proposed to overcome the limitations of Context tree weighting (CTW) method, which applied in CABAC know as Context Adaptive Binary Arithmetic Coding. This CABAC uses KT estimators and relies on beginning with the Bayesian approach to determine the true distribution of the next symbol to select for data compression. This approach is suitable only if the true distribution is stationary, the proposed SANDCTW that uses discounted KT estimators, which is suitable if the distribution is non-stationary and it reduces the computation and memory cost. Additionally, Block size sustained intra mode detection (BSSIMD) is proposed based on the mass-center and sub-sampling approach. In this approach, all correlation directions about the entire block associated to the intra-prediction mode and DC mode directions determined by using mass-center vector. Then, the modes corresponding to the determined directions selected as the best intra-prediction candidates during the intra-coding process for computing Rate-Distortion Optimization (RDO) with less complexity. The bit rate of 60–70 frames per second (fps) achieved in this technique based on the different block, size. Thus, the bit rate is also reduced significantly compared with the preceding H.265 standard.