Národní úložiště šedé literatury Nalezeno 2 záznamů.  Hledání trvalo 0.00 vteřin. 
Bayesian Entropic Evolution
Gottvald, Aleš
We develop a quantitative and experimantally testable theory of evolution, based on Bayesian and Entropic concepts. Probability and information are given central roles in evolutionary processes. Substantial evidence is now available that all logically consistent probabilistic transformations, in evolutionary processes or elsewhere, obey fundamental laws of Bayesian Probability Theory. In particular, Bayes' Theorem and the maximum Entropy Principle imply many quantitative and testable consequences also in evolutionary processes, in full analogy with Statistical Mechanics. Every evolutionary process may be treated as a chain of Bayesian probabilistic interfereces from incomplete information. Only systems equipped with a Bayesian processing layer react most rationally to a changing information environment, which brings some evolutionary advantages.
Evolution and Genetic Optimization of Higher-Order Shim Coils for NMR
Chládek, Jan ; Konzbul, Pavel ; Ošmera, P.
Evolutionary and genetic scholastic optimizations were explored for designing NMR gradient and shim coils. Some hierarchical self-adaptive algorithms, based on a concept of meta-optimization, were tested. Time-demands and quality of the solutions were compared for various optimisers. A broad range of contrasting features was found for different settings of governing parameters and for different constraint handling's. Therefore a fuzzy logic in meta-decision level and an adaptative barrier function were used for optimization of multi-modal objective function. The using of meta-level optimization of different parameters was identified as the major challenge for future investigations and improvements of the optimizers.

Chcete být upozorněni, pokud se objeví nové záznamy odpovídající tomuto dotazu?
Přihlásit se k odběru RSS.