|
|
The Impact of Candidate Solution Mappings on Evolutionary Algorithm Efficiency
Hrbáček, Jiří ; Korček, Pavol (referee) ; Křivánek, Jan (advisor)
The Concern of the present study is summarizing knowledges in the theory of mapping candidate solutions , analysis and application of evolutionary algorithms. The study provides summary of the evolutionary algorithms, classification and application. The target of the study is links gained knowledge from sectionS of ; evolutionary algorithms, mapping candidate solutions and creations of a system that will demonstrate and influence mapping the efficiency of the evolutionary algorithms succesfully.
|
|
|
Functional Annotation of Nucleotide Polymorphism Using Evolution Strategy
Šalanda, Ondřej ; Martínek, Tomáš (referee) ; Bendl, Jaroslav (advisor)
This thesis brings a new approach to the prediction of the the effect of amino acid substitution. The main goal is to create a new meta-tool, which combines evaluations of eight already implemented prediction tools. The use of weighted consensus over those tools should lead to better accuracy and versatility of prediction. The novelty of developed tool lies in involving evolution strategy with experimentally defined parameters as a way to determine the best weight distribution. At the end, a complex comparison and evaluation of results is given.
|
|
|
Prediction of Protein Stability upon Mutations Using Evolution Strategy
Pavlík, David ; Martínek, Tomáš (referee) ; Bendl, Jaroslav (advisor)
This master's thesis deals with the matter of predicting the effects of aminoacid substitutions on protein stability. The main aim is to design meta-classifier that combines the results of the selected prediction tools. An evolution strategy was used to find the best weights for each of the selected tools with the aim of achieving better prediction performance compared to that achieved by using these tools separately. Five different and obtainable prediction tools were selected and their prediction outputs were weighted. Two different approaches of evolution strategy are investigated and compared: evolution strategy with the 1/5-rule and evolution strategy with the type 2 of control parameters self-adaptation. Two independent datasets of mutations were created for training and evaluating the performance of designed meta-classifier. The performed experiments and obtained results suggest that the evolution strategy could be considered as a~beneficial approach for prediction of protein stability changes. However, the special attention must be paid to careful selection of tools for integration and compilation of training and testing datasets.
|
|
|
Optimization of Aircraft Tracker Parameters
Samek, Michal ; Vlk, Jan (referee) ; Smrž, Pavel (advisor)
Diplomová práce se zabývá optimalizací systému pro sledování letadel, využívaného pro řízení letového provozu. Je popsána metodika vyhodnocování přesnosti sledovacího systému a přehled relevantních algoritmů pro sledování objektů. Dále jsou navrženy tři přístupy k řešení problému. První se pokouší identifikovat parametry filtrovacích algoritmů pomocí algoritmu Expectation-Maximisation, implementací metody maximální věrohodnosti. Druhý přístup je založen na prostých odhadech parametrů normálního rozložení z naměřených a referenčních dat. Nakonec je zkoumána možnost řešení pomocí optimalizačního algoritmu Evoluční strategie. Závěrečné vyhodnocení ukazuje, že třetí přístup je pro daný problém nejvhodnější.
|
|
|
Prediction of Protein Stability upon Amino Acid Mutations Using Evolution Strategy
Kadlec, Miroslav ; Burgetová, Ivana (referee) ; Bendl, Jaroslav (advisor)
This thesis is focused on predicting the impact of amino acid substitution on protein stability. The main goal is to create a consensual predictor that uses the outputs of chosen existing tools in order to improve accuracy of prediction. The optimal consensus of theese tools was designed using evolution strategies in three variants: 1/5 success rule, self-adaptation variant and the CMA-ES method. Then, the quality of calculated weight vectors was tested on the independent dataset. Although the highest prediction performance was attained by self-adaptation method, the differences between all three variants were not significant. Compared to the individual tools, the predictions provided by consensual methods were generally more accurate - the self-adaptation variant imporved the Pearson's corelation coeficient of the predictions by 0,057 on the training dataset. On the testing dataset, the improvement of designed method was smaller (0,040). Relatively low improvement of prediction performance (both on the training and the testing dataset) were caused by the fact, that for some records of testing dataset, some individual tools vere not able to provide their results. When omitting these records, consensual method improved the Pearson's corelations coeficient by 0,118.
|
|
|
Neural networks and evolutionary algorithms
Vágnerová, Jitka ; Rychtárik, Milan (referee) ; Hrubeš, Jan (advisor)
Objective of this master's thesis is optimizing of neral network topology using some of evolutionary algorithms. The backpropagation neural network was optimized using genetic algorithms, evolutionary programming and evolutionary strategies. The text contains an application in the Matlab environment which applies these methods to simple tasks as pattern recognition and function prediction. Created graphs of fitness and error functions are included as a result of this thesis.
|
guest ::
