Improving time series forecastingaccuracy is an important yet often difficult task.Both theoretical and empirical findings haveindicated that integration of several models is an effectiveway to improve predictive performance, especiallywhen the models in combination are quite different. In this paper,a model of the hybrid artificial neural networks andfuzzy model is proposed for time series for...

Nowadays, algorithms with fast convergence, small memory footprints, and low per-iteration complexity are particularly favorable for artificial intelligence applications. In this paper, we propose a doubly stochastic algorithm with a novel accelerating multi-momentum technique to solve large scale empirical risk minimization problem for learning tasks. While enjoying a provably superior converg...

We proposed a doubly stochastic primal-dual coordinate optimization algorithm for regularized empirical risk minimization that can be formulated as a saddlepoint problem. Different from existing coordinate methods, the proposed method randomly samples both primal and dual coordinates to update solutions, which is a desirable property when applied to data with both a high dimension and a large s...

this article describes a knowledge management (km) success model that is derived from observations generated through an empirical study of km in an iranian consultant company and km success factors in the literature which is shown that company has a weak situation in each factor. so company needs a measurement model in order to assess knowledge management implementation processes in each step t...

This study describes an EOF-based technique for statistical downscaling of high spatial resolution monthly mean precipitation from large scale reanalysis circulation fields. The method is demonstrated and evaluated for four widely-separated locations: the Southeast United States, the Upper Colorado River Basin, China’s Jiangxi Province, and Central Europe. For each location, the EOF-based downs...

The success of regularized risk minimization approaches to classification with linear models depends crucially on the selection of a regularization term that matches with the learning task at hand. If the necessary domain expertise is rare or hard to formalize, it may be difficult to find a good regularizer. On the other hand, if plenty of related or similar data is available, it is a natural a...

This paper investigates the performance of efficient portfolios in a financial market with heterogeneous investors including rational traders, noise traders, and chartists. A generalization of the security market line result states that, regardless of the diversity of beliefs, the portfolios of rational investors with mean-variance preferences are mean-variance efficient in the sense of classic...

An ice core extracted from Holtedahlfonna ice cap, western Spitsbergen, record spanning the period 1700–2005, was analyzed for major ions. The leading empirical orthogonal function (EOF) component is correlated with an index of summer melt (log([Na]/[Mg]) from 1850 and shows that almost 50% of the variance can be attributed to seasonal melting since the beginning of the industrial revolution. T...

We address the problem of algorithmic fairness: ensuring that sensitive variables do not unfairly influence the outcome of a classifier. We present an approach based on empirical risk minimization, which incorporates a fairness constraint into the learning problem. It encourages the conditional risk of the learned classifier to be approximately constant with respect to the sensitive variable. W...

Theoretical available values of thermal conductivity are commonly different from the experimentally obtained values. In the present work, a thermodynamic system has been considered and variations in temperature in different elements were analyzed and corresponding thermal conductivities were calculated. These values were compared and the corrected values were justified on the basis of a mathema...