On Structure of Predictors Allowing Distributed Dynamic Bayesian Decision-Making
Šmídl, Václav
2007 - anglický Decentralized adaptive control is based on the use of many local controllers in parallel, each of them estimating its own local model and pursuing its local aims. If each controller designs its strategy using only its own model, the resulting control may be poor since consequences of actions of the neighbors are not taken into account. We seek a way how to improve algorithm of decision strategy design of a single local controller without significant increase in complexity of the local model or complexity of the design procedure. In this paper we study variants of distributed dynamic programming that could be evaluated locally. Specifically, we will investigate variants of the fully probabilistic control strategy design. Distributed and cen- tralized control strategies will be compared. Distribuované adaptivní řízení spočívá v paralelním běhu několika adaptivních regulátorů, z nichž každý odhaduje svůj model a navrhuje valstní strategii řízení. Pokud by takový regulátor nevzal do úvahy přítomnost ostatních regulátorů, může dojít k nežádoucímu chování. V této práci se zabýváme Bayesovským přístupem k problému. Cílem je zjistit jaká struktura centralizovaného prediktoru umožňuje návrh distribuované strategie. Klíčová slova: distributed control; Bayesian decision-making; decentralized adaptive control; dynamic programming Plné texty jsou dostupné na jednotlivých ústavech Akademie věd ČR. On Structure of Predictors Allowing Distributed Dynamic Bayesian Decision-Making

Decentralized adaptive control is based on the use of many local controllers in parallel, each of them estimating its own local model and pursuing its local aims. If each controller designs its ...

Šmídl, Václav
Ústav teorie informace a automatizace, 2007

Decision Making with Imperfect Knowledge Accumulation
Kárný, Miroslav
2007 - anglický Formal basis of Bayesian decision making with non-expanding knowledge accumulation is proposed. Je navržen formální základ bayesovského rozhodování při nenarůstajícím rozsahu znalostí Klíčová slova: Bayesian decision making; knowledge accumulation Plné texty jsou dostupné na jednotlivých ústavech Akademie věd ČR. Decision Making with Imperfect Knowledge Accumulation

Formal basis of Bayesian decision making with non-expanding knowledge accumulation is proposed.

Kárný, Miroslav
Ústav teorie informace a automatizace, 2007

An Adaptive Feed-Forward Control
Kárný, Miroslav
2007 - anglický Feed-forward controllers are important in a range of control problems. Their importance is obvious in tasks in which potential of feedback control is limited, for instance, in systems with long transportation delays. Formally, their design can be approached by a standard methodology of optimal stochastic control. It can be consistently performed by combining Bayesian learning and dynamic programming. This formal solution can, however, rarely be converted into a computationally feasible algorithms. Thus, various approximations are searched for. The current report deals with a specific type of approximation based on a projection of optimal /emph{anticipating} control strategy to a non-anticipating one. This approximation way suits to the feed-forward control in which the selected system inputs influence the state of the controller but not the system-related data used in the feed-forward loop. V práci je popsán speciální návrh adaptivní dopředné vazby založený na neanticipativní aproximaci optimální anticipativní strategie. Klíčová slova: adaptive feed=forward control; approximation Plné texty jsou dostupné na jednotlivých ústavech Akademie věd ČR. An Adaptive Feed-Forward Control

Feed-forward controllers are important in a range of control problems. Their importance is obvious in tasks in which potential of feedback control is limited, for instance, in systems with long ...

Kárný, Miroslav
Ústav teorie informace a automatizace, 2007

Image Segmentation Based on Local Noise Variance
Saic, Stanislav; Mahdian, Babak
2007 - anglický New segmentation method detecting changes in noise variance is introduced. Several examples are shown to demonstrate the method’s output. Je předvedena nová metoda pro detekci zmen variance sumu v digitálním obraze. Na řadě příkladů jsoou uvedeny výstupy metody. Klíčová slova: Digital image; Segmentation; Noise; Tampering Plné texty jsou dostupné na jednotlivých ústavech Akademie věd ČR. Image Segmentation Based on Local Noise Variance

New segmentation method detecting changes in noise variance is introduced. Several examples are shown to demonstrate the method’s output....

Saic, Stanislav; Mahdian, Babak
Ústav teorie informace a automatizace, 2007

Spectral gap for zero processes on graphs
Fajfrová, Lucie
2007 - anglický The paper presents a method how the problem of finding a lower bound on the spectral gap for a zero range process on a connected graph could be transformed into the problem of finding a lower bound on spectral gap of a simple random walk on the graph. V clanku se ukazuje, jak lze ulohu nalezeni odhadu spektralni mezery pro Zero-range proces na grafu prevest na ulohu hledani spektralni mezery nahodne prochazky na grafu Klíčová slova: particle systems; zero range dynamics; spectral gap Plné texty jsou dostupné na jednotlivých ústavech Akademie věd ČR. Spectral gap for zero processes on graphs

The paper presents a method how the problem of finding a lower bound on the spectral gap for a zero range process on a connected graph could be transformed into the problem of finding a lower ...

Fajfrová, Lucie
Ústav teorie informace a automatizace, 2007

Convexity and dependence in chance-constrained programming
Houda, Michal
2007 - anglický Klíčová slova: stochastich programmnig; weak dependence; stability Plné texty jsou dostupné na jednotlivých ústavech Akademie věd ČR. Convexity and dependence in chance-constrained programming

Houda, Michal
Ústav teorie informace a automatizace, 2007

Empirical Estimates via Stability in Stochastic Programming
Kaňková, Vlasta
2007 - anglický It is known that optimization problems depending on a probability measure correspond to many applications. It is also known that these problems belong mostly to a class of nonlinear optimization problems and, moreover, that very often an ``underlying" probability measure is not completely known. The aim of the research report is to deal with the case when an empirical measure substitutes the theoretical one. In particular, the aim is to generalize reults dealing with convergence rate in the case of empirical esrimates. The introduced results are based on the stability results corresponding to the Wasserstein metric. A relationship berween tails of one-dimensional marginal distribution functions and exponentional rate of convergence are introduced. The corresponding results are focus mainly on ``classical" type of problems corresponding to the cases with penalty and recourse. However, an integer simple recourse case and some special risk funkcionals are discussed also. Klíčová slova: Stochastic programming; stability; Wasserstein metric; empirical estimates; convergence rate; problems with penalty and recourse; integer simple recourse case; resk funkcionals Plné texty jsou dostupné na jednotlivých ústavech Akademie věd ČR. Empirical Estimates via Stability in Stochastic Programming

It is known that optimization problems depending on a probability measure correspond to many applications. It is also known that these problems belong mostly to a class of nonlinear optimization ...

Kaňková, Vlasta
Ústav teorie informace a automatizace, 2007

O divergenci a fluktuaci proměnných veličin a pravděpodobnostních distribucí
Vajda, Igor
2007 - český O divergenci a fluktuaci proměnných veličin a pravděpodobnostních distribucí Klíčová slova: divergence; probability measure Plné texty jsou dostupné na jednotlivých ústavech Akademie věd ČR. O divergenci a fluktuaci proměnných veličin a pravděpodobnostních distribucí

O divergenci a fluktuaci proměnných veličin a pravděpodobnostních distribucí

Vajda, Igor
Ústav teorie informace a automatizace, 2007

Performance Analysis of Extended EFICA Algorithm
Koldovský, Zbyněk; Málek, J.; Tichavský, Petr; Yannick, D.; Shahram, H.
2007 - anglický This paper supports the document "Extension of EFICA Algorithm for Blind Separation of Piecewise Stationary Non Gaussian Sources." Tento clanek je doplnkem k clanku "Extension of EFICA Algorithm for Blind Separation of Piecewise Stationary Non Gaussian Sources." Klíčová slova: Cramer-Rao Lower Bound; Independent Component Analysis; Blind Source Separation; EFICA Algorithm Plné texty jsou dostupné na jednotlivých ústavech Akademie věd ČR. Performance Analysis of Extended EFICA Algorithm

This paper supports the document "Extension of EFICA Algorithm for Blind Separation of Piecewise Stationary Non Gaussian Sources."...

Koldovský, Zbyněk; Málek, J.; Tichavský, Petr; Yannick, D.; Shahram, H.
Ústav teorie informace a automatizace, 2007

Predictive Monitoring of Radiation Situation
Nedoma, Petr; Pecha, Petr; Kuča, P.
2007 - anglický Report describes proposition how predictive monitoring of radiological situation from measurements of Early Warning Network of the Czech Republic could be realised. The presented results indicate that mixtures of auto-regresive models improve predictions comparing to plain auto-regresion. Zpráva popisuje návrh prediktivního monitorování radiační situace z měření přicházejících z radiační sítě včasného zjištění ČR. Uvedené výsledky dokládají, že použité směsi auto-regresivních modelů zlepšují predikční schopnost ve srovnání s jednoduchou auto-regresí. Klíčová slova: Monitoring; accidents; warning; protection Plné texty jsou dostupné na jednotlivých ústavech Akademie věd ČR. Predictive Monitoring of Radiation Situation

Report describes proposition how predictive monitoring of radiological situation from measurements of Early Warning Network of the Czech Republic could be realised. The presented results indicate that ...

Nedoma, Petr; Pecha, Petr; Kuča, P.
Ústav teorie informace a automatizace, 2007