A Modified Limited-Memory BNS Method for Unconstrained Minimization Derived from the Conjugate Directions Idea
Vlček, Jan; Lukšan, Ladislav
2015 - anglický A modification of the limited-memory variable metric BNS method for large scale unconstrained optimization of the differentiable function $f:{\cal R}^N\to\cal R$ is considered, which consists in corrections (based on the idea of conjugate directions) of difference vectors for better satisfaction of the previous quasi-Newton conditions. In comparison with [11], more previous iterations can be utilized here. For quadratic objective functions, the improvement of convergence is the best one in some sense, all stored corrected difference vectors are conjugate and the quasi-Newton conditions with these vectors are satisfied. The algorithm is globally convergent for convex sufficiently smooth functions and our numerical experiments indicate its efficiency. Klíčová slova: large scale unconstrained optimization; numerical experiments; limited-memory variable metric method; BNS method; quasi-Newton method; convergence Plné texty jsou dostupné v digitálním repozitáři Akademie Věd. A Modified Limited-Memory BNS Method for Unconstrained Minimization Derived from the Conjugate Directions Idea

A modification of the limited-memory variable metric BNS method for large scale unconstrained optimization of the differentiable function $f:{\cal R}^N\to\cal R$ is considered, which consists in ...

Vlček, Jan; Lukšan, Ladislav
Ústav informatiky, 2015

On Two Methods for the Parameter Estimation Problem with Spatio-Temporal FRAP Data
Papáček, Š.; Jablonský, J.; Matonoha, Ctirad
2015 - anglický FRAP (Fluorescence Recovery After Photobleaching) is a measurement technique for determination of the mobility of fluorescent molecules (presumably due to the diffusion process) within the living cells. While the experimental setup and protocol are usually fixed, the method used for the model parameter estimation, i.e. the data processing step, is not well established. In order to enhance the quantitative analysis of experimental (noisy) FRAP data, we firstly formulate the inverse problem of model parameter estimation and then we focus on how the different methods of data pre- processing influence the confidence interval of the estimated parameters, namely the diffusion constant $p$. Finally, we present a preliminary study of two methods for the computation of a least-squares estimate $\hat{p}$ and its confidence interval. Klíčová slova: parameter estimation; fluorescence recovery after photobleaching; diffusion equation; Moullineaux method; Fisher information matrix; sensitivity analysis; confidence intervals; uncertainty quantification Plné texty jsou dostupné v digitálním repozitáři Akademie Věd. On Two Methods for the Parameter Estimation Problem with Spatio-Temporal FRAP Data

FRAP (Fluorescence Recovery After Photobleaching) is a measurement technique for determination of the mobility of fluorescent molecules (presumably due to the diffusion process) within the living ...

Papáček, Š.; Jablonský, J.; Matonoha, Ctirad
Ústav informatiky, 2015

Measures for Classification Results Evaluation
Řezanková, Hana; Húsek, Dušan
2015 - anglický Klíčová slova: similarity measures; measures of agreement; success rate of classification Plné texty jsou dostupné v digitálním repozitáři NUŠL Measures for Classification Results Evaluation

Řezanková, Hana; Húsek, Dušan
Ústav informatiky, 2015

Causation Entropy Principle and Bayesian Inference to Causal Networks
Coufal, David; Hlinka, Jaroslav
2015 - anglický Klíčová slova: causal links; causation entropy; Bayesian inference Plné texty jsou dostupné v digitálním repozitáři NUŠL Causation Entropy Principle and Bayesian Inference to Causal Networks

Coufal, David; Hlinka, Jaroslav
Ústav informatiky, 2015

IINC Software
Jiřina, Marcel
2015 - anglický Klíčová slova: Multidimensional data; Classifier; Distance; Metrics; Hassanat metrics; k-NN; IINC Plné texty jsou dostupné v digitálním repozitáři NUŠL IINC Software

Jiřina, Marcel
Ústav informatiky, 2015

Probabilistic learning model PAC - lecture notes
Hakl, František
2015 - anglický Klíčová slova: supervised learning; PAC model; Vapnik-Chervonenkis dimension; learning algorithms Plné texty jsou dostupné v digitálním repozitáři NUŠL Probabilistic learning model PAC - lecture notes

Hakl, František
Ústav informatiky, 2015

Indecisive Belief Functions
Daniel, Milan
2015 - anglický This study presents an idea of indecisive functions, their general and also special definitions, plausibility and pignistic indecisive belief functions. The rich structure of indecisive belief functions is studied in general, and also in special views: both general substructures and indecisive belief functions on three-element and general finite frames of discernment. We are focused to pignistic and contour (plausibility) indecisive belief functions, including their mutual relationship in our study. The later have interesting algebraic structure related to Dempster’s rule of combination. Klíčová slova: belief function; theory of evidence; Dempster-Shafer theory; Dempster’s semigroup Dokument je dostupný na externích webových stránkách. Indecisive Belief Functions

This study presents an idea of indecisive functions, their general and also special definitions, plausibility and pignistic indecisive belief functions. The rich structure of indecisive belief ...

Daniel, Milan
Ústav informatiky, 2015

Synergy between the Parameter Estimation and a Design Variable Optimization for FRAP Experiments
Matonoha, Ctirad; Papáček, Š.
2015 - anglický Plné texty jsou dostupné v digitálním repozitáři Akademie Věd. Synergy between the Parameter Estimation and a Design Variable Optimization for FRAP Experiments

Matonoha, Ctirad; Papáček, Š.
Ústav informatiky, 2015

Dynamic Contact Problems in Bone Neoplasm Analyses and the Primal-Dual Active Set (PDAS) Method
Nedoma, Jiří
2015 - anglický In the contribution growths of the neoplasms (benign and malignant tumors and cysts), located in a system of loaded bones, will be simulated. The main goal of the contribution is to present the useful methods and efficient algorithms for their solutions. Because the geometry of the system of loaded and possible fractured bones with enlarged neoplasms changes in time, the corresponding mathematical models of tumor’s and cyst’s evolutions lead to the coupled free boundary problems and the dynamic contact problems with or without friction. The discussed parts of these models will be based on the theory of dynamic contact problems without or with Tresca or Coulomb frictions in the visco-elastic rheology. The numerical solution of the problem with Coulomb friction is based on the semi-implicit scheme in time and the finite element method in space, where the Coulomb law of friction at every time level will be approximated by its value from the previous time level. The algorithm for the corresponding model of friction will be based on the discrete mortar formulation of the saddle point problem and the primal-dual active set algorithm. The algorithm for the Coulomb friction model will be based on the fixpoint algorithm, that will be an extension of the PDAS algorithm for the Tresca friction. In this algorithm the friction bound is iteratively modified using the normal component of the Lagrange multiplier. Thus the friction bound and the active and inactive sets are updated in every step of the iterative algorithm and at every time step corresponding to the semi-implicit scheme. Klíčová slova: dynamic contact problems; mathematical models of neoplasms - tumors and cysts; Coulomb and Tresca frictions; variational formulation; semi-implicit scheme; FEM; mortar approximation; PDAS algorithm Dokument je dostupný na externích webových stránkách. Dynamic Contact Problems in Bone Neoplasm Analyses and the Primal-Dual Active Set (PDAS) Method

In the contribution growths of the neoplasms (benign and malignant tumors and cysts), located in a system of loaded bones, will be simulated. The main goal of the contribution is to present the useful ...

Nedoma, Jiří
Ústav informatiky, 2015

Homomorphic Coordinates of Dempster’s Semigroup
Daniel, Milan
2015 - anglický Coordinates of belief functions on two-element frame of discernment are defined using homomorphisms of Dempster’s semigroup (the algebra of belief functions with Dempster’s rule). Three systems of the coordinates (h-f, h-f0, and coordinates based on decomposition of belief functions) are analysed with a focus to their homomorphic properties. Further, ideas of generalisation of the investigated systems of coordinates to general finite frame of discernment are presented. Klíčová slova: belief functions; Dempster-Shafer Theory; Dempster's semigroup; homomorphism; homomorphic coordinates Dokument je dostupný na externích webových stránkách. Homomorphic Coordinates of Dempster’s Semigroup

Coordinates of belief functions on two-element frame of discernment are defined using homomorphisms of Dempster’s semigroup (the algebra of belief functions with Dempster’s rule). Three systems of the ...

Daniel, Milan
Ústav informatiky, 2015