The Computational Power of Neural Networks and Representations of Numbers in Non-Integer Bases.
Šíma, Jiří
2017 - anglický We briefly survey the basic concepts and results concerning the computational power of neural networks which basically depends on the information content of weight parameters. In particular, recurrent neural networks with integer, rational, and arbitrary real weights are classified within the Chomsky and finer complexity hierarchies. Then we refine the analysis between integer and rational weights by investigating an intermediate model of integer-weight neural networks with an extra analog rational-weight neuron (1ANN). We show a representation theorem which characterizes the classification problems solvable by 1ANNs, by using so-called cut languages. Our analysis reveals an interesting link to an active research field on non-standard positional numeral systems with non-integer bases. Within this framework, we introduce a new concept of quasi-periodic numbers which is used to classify the computational power of 1ANNs within the Chomsky hierarchy. Klíčová slova: neural network; Chomsky hierarchy; beta-expansion; cut language Plné texty jsou dostupné na jednotlivých ústavech Akademie věd ČR. The Computational Power of Neural Networks and Representations of Numbers in Non-Integer Bases.

We briefly survey the basic concepts and results concerning the computational power of neural networks which basically depends on the information content of weight parameters. In particular, recurrent ...

Šíma, Jiří
Ústav informatiky, 2017

An adaptive recursive multilevel approximate inverse preconditioning: Computation of the Schur complement
Kopal, Jiří; Rozložník, Miroslav; Tůma, Miroslav
2017 - anglický Plné texty jsou dostupné v digitálním repozitáři Akademie Věd. An adaptive recursive multilevel approximate inverse preconditioning: Computation of the Schur complement

Kopal, Jiří; Rozložník, Miroslav; Tůma, Miroslav
Ústav informatiky, 2017

Exact Inference In Robust Econometrics under Heteroscedasticity
Kalina, Jan; Peštová, Barbora
2017 - anglický The paper is devoted to the least weighted squares estimator, which is one of highly robust estimators for the linear regression model. Novel permutation tests of heteroscedasticity are proposed. Also the asymptotic behavior of the permutation test statistics of the Goldfeld-Quandt and Breusch-Pagan tests is investigated. A numerical experiment on real economic data is presented, which also shows how to perform a robust prediction model under heteroscedasticity. Theoretical results may be simply extended to the context of multivariate quantiles Klíčová slova: heteroscedasticity; robust statistics; regression; diagnostic tools; economic data Dokument je dostupný na externích webových stránkách. Exact Inference In Robust Econometrics under Heteroscedasticity

The paper is devoted to the least weighted squares estimator, which is one of highly robust estimators for the linear regression model. Novel permutation tests of heteroscedasticity are proposed. Also ...

Kalina, Jan; Peštová, Barbora
Ústav informatiky, 2017

On the Optimization of Initial Conditions for a Model Parameter Estimation
Matonoha, Ctirad; Papáček, Š.; Kindermann, S.
2017 - anglický The design of an experiment, e.g., the setting of initial conditions, strongly influences the accuracy of the process of determining model parameters from data. The key concept relies on the analysis of the sensitivity of the measured output with respect to the model parameters. Based on this approach we optimize an experimental design factor, the initial condition for an inverse problem of a model parameter estimation. Our approach, although case independent, is illustrated at the FRAP (Fluorescence Recovery After Photobleaching) experimental technique. The core idea resides in the maximization of a sensitivity measure, which depends on the initial condition. Numerical experiments show that the discretized optimal initial condition attains only two values. The number of jumps between these values is inversely proportional to the value of a diffusion coefficient D (characterizing the biophysical and numerical process). The smaller value of D is, the larger number of jumps occurs. Klíčová slova: FRAP; sensitivity analysis; optimal experimental design; parameter estimation; finite differences Plné texty jsou dostupné v digitálním repozitáři Akademie Věd. On the Optimization of Initial Conditions for a Model Parameter Estimation

The design of an experiment, e.g., the setting of initial conditions, strongly influences the accuracy of the process of determining model parameters from data. The key concept relies on the analysis ...

Matonoha, Ctirad; Papáček, Š.; Kindermann, S.
Ústav informatiky, 2017

The Computational Power of Neural Networks and Representations of Numbers in Non-Integer Bases
Šíma, Jiří
2017 - anglický We briefly survey the basic concepts and results concerning the computational power of neural networks which basically depends on the information content of weight parameters. In particular, recurrent neural networks with integer, rational, and arbitrary real weights are classified within the Chomsky and finer complexity hierarchies. Then we refine the analysis between integer and rational weights by investigating an intermediate model of integer-weight neural networks with an extra analog rational-weight neuron (1ANN). We show a representation theorem which characterizes the classification problems solvable by 1ANNs, by using so-called cut languages. Our analysis reveals an interesting link to an active research field on non-standard positional numeral systems with non-integer bases. Within this framework, we introduce a new concept of quasi-periodic numbers which is used to classify the computational power of 1ANNs within the Chomsky hierarchy. Klíčová slova: neural network; Chomsky hierarchy; beta-expansion; cut language Plné texty jsou dostupné na jednotlivých ústavech Akademie věd ČR. The Computational Power of Neural Networks and Representations of Numbers in Non-Integer Bases

We briefly survey the basic concepts and results concerning the computational power of neural networks which basically depends on the information content of weight parameters. In particular, recurrent ...

Šíma, Jiří
Ústav informatiky, 2017

On the optimal initial conditions for an inverse problem of model parameter estimation
Matonoha, Ctirad; Papáček, Š.
2017 - anglický Plné texty jsou dostupné v digitálním repozitáři Akademie Věd. On the optimal initial conditions for an inverse problem of model parameter estimation

Matonoha, Ctirad; Papáček, Š.
Ústav informatiky, 2017

A Generalized Limited-Memory BNS Method Based on the Block BFGS Update
Vlček, Jan; Lukšan, Ladislav
2017 - anglický A block version of the BFGS variable metric update formula is investigated. It satisfies the quasi-Newton conditions with all used difference vectors and gives the best improvement of convergence in some sense for quadratic objective functions, but it does not guarantee that the direction vectors are descent for general functions. To overcome this difficulty and utilize the advantageous properties of the block BFGS update, a block version of the limited-memory BNS method for large scale unconstrained optimization is proposed. The algorithm is globally convergent for convex sufficiently smooth functions and our numerical experiments indicate its efficiency. Klíčová slova: unconstrained minimization; block variable metric methods; limited-memory methods; the BFGS update; global convergence; numerical results Plné texty jsou dostupné v digitálním repozitáři Akademie Věd. A Generalized Limited-Memory BNS Method Based on the Block BFGS Update

A block version of the BFGS variable metric update formula is investigated. It satisfies the quasi-Newton conditions with all used difference vectors and gives the best improvement of convergence in ...

Vlček, Jan; Lukšan, Ladislav
Ústav informatiky, 2017

The use of sequential quadratic programming for solving reachability problems
Kuřátko, Jan
2017 - anglický Plné texty jsou dostupné v digitálním repozitáři Akademie Věd. The use of sequential quadratic programming for solving reachability problems

Kuřátko, Jan
Ústav informatiky, 2017

Robust Regression Estimators: A Comparison of Prediction Performance
Kalina, Jan; Peštová, Barbora
2017 - anglický Regression represents an important methodology for solving numerous tasks of applied econometrics. This paper is devoted to robust estimators of parameters of a linear regression model, which are preferable whenever the data contain or are believed to contain outlying measurements (outliers). While various robust regression estimators are nowadays available in standard statistical packages, the question remains how to choose the most suitable regression method for a particular data set. This paper aims at comparing various regression methods on various data sets. First, the prediction performance of common robust regression estimators are compared on a set of 24 real data sets from public repositories. Further, the results are used as input for a metalearning study over 9 selected features of individual data sets. On the whole, the least trimmed squares turns out to be superior to the least squares or M-estimators in the majority of the data sets, while the process of metalearning does not succeed in a reliable prediction of the most suitable estimator for a given data set. Klíčová slova: robust estimation; linear regression; prediction; outliers; metalearning Plné texty jsou dostupné na jednotlivých ústavech Akademie věd ČR. Robust Regression Estimators: A Comparison of Prediction Performance

Regression represents an important methodology for solving numerous tasks of applied econometrics. This paper is devoted to robust estimators of parameters of a linear regression model, which are ...

Kalina, Jan; Peštová, Barbora
Ústav informatiky, 2017

Various Approaches to Szroeter’s Test for Regression Quantiles
Kalina, Jan; Peštová, Barbora
2017 - anglický Regression quantiles represent an important tool for regression analysis popular in econometric applications, for example for the task of detecting heteroscedasticity in the data. Nevertheless, they need to be accompanied by diagnostic tools for verifying their assumptions. The paper is devoted to heteroscedasticity testing for regression quantiles, while their most important special case is commonly denoted as the regression median. Szroeter’s test, which is one of available heteroscedasticity tests for the least squares, is modified here for the regression median in three different ways: (1) asymptotic test based on the asymptotic representation for regression quantiles, (2) permutation test based on residuals, and (3) exact approximate test, which has a permutation character and represents an approximation to an exact test. All three approaches can be computed in a straightforward way and their principles can be extended also to other heteroscedasticity tests. The theoretical results are expected to be extended to other regression quantiles and mainly to multivariate quantiles. Klíčová slova: Heteroscedasticity; Regression median; Diagnostic tools; Asymptotics Plné texty jsou dostupné na vyžádání prostřednictvím repozitáře Akademie věd. Various Approaches to Szroeter’s Test for Regression Quantiles

Regression quantiles represent an important tool for regression analysis popular in econometric applications, for example for the task of detecting heteroscedasticity in the data. Nevertheless, they ...

Kalina, Jan; Peštová, Barbora
Ústav informatiky, 2017