|
|
Evolutionary Design Using Rewriting Systems
Nétková, Barbora ; Hyrš, Martin (referee) ; Bidlo, Michal (advisor)
This master’s thesis proposes a method for the evolutionary design of rewriting systems. In particular, genetic algorithm will be applied to design rewriting rules for a specific variant of Lindenmayer system. The evolved rules of such grammar will be applied to generate growing sorting networks. Some distinct approaches to the rewriting process and construction of the sorting networks will be investigated. It will be shown that the evolution is able to successfully design rewriting rules for the proposed variants of rewriting processes. The results obtained exhibit abilities to successfully create partially growing sorting networks, which was evolved to grow for fewer inputs and in subsequent iterations grows up to 36 inputs.
|
|
|
Evolution algorithms for ultrasound perfusion analysis
Kolářová, Jana ; Odstrčilík, Jan (referee) ; Mézl, Martin (advisor)
This master´s thesis is focused on the application of evolutionary algorithms for interleaving data obtained by ultrasound scanning of tissue. The interleaved curve serves to estimate perfusion parameters, thus allowing to detect possible pathophysiology in the scanned area. The theoretical introduction is devoted to perfusion and its parameters, contrast agents for ultrasonic application, ultrasonic modality scanning, optimization, evolutionary algorithms in general and two selected evolutionary algorithms - genetic algorithm and bee algorithm. These algorithms were tested on noisy data obtained from clinical images of mice with tumor. The final part summarizes the results of the practical part and provides suggestions and recommendations for further possible development.
|
|
|
Automated Design Methodology for Approximate Low Power Circuits
Mrázek, Vojtěch ; Bosio, Alberto (referee) ; Fišer, Petr (referee) ; Sekanina, Lukáš (advisor)
Rozšiřování moderních vestavěných a mobilních systémů napájených bateriemi zvyšuje požadavky na návrh těchto systémů s ohledem na příkon. Přestože moderní návrhové techniky optimalizují příkon, elektrická spotřeba těchto obvodů stále roste díky jejich složitosti. Nicméně existuje celá řada aplikací, kde nepotřebujeme získat úplně přesný výstup. Díky tomu se objevuje technika zvaná aproximativní (přibližné) počítání, která umožňuje za cenu zanesení malé chyby do výpočtu významně redukovat příkon obvodů. V práci se zaměřujeme na použití evolučních algoritmů v této oblasti. Ačkoliv již tyto algoritmy byly úspěšně použity v syntéze přesných i aproximativních obvodů, objevují se problémy škálovatelnosti - schopnosti aproximovat složité obvody. Cílem této disertační práce je ukázat, že aproximační logická syntéza založená na genetickém programování umožňuje dosáhnout vynikajícího kompromisu mezi spotřebou a chybou. Byla provedena analýza čtyř různých aplikacích na třech úrovních popisu. Pomocí kartézského genetického programování s modifikovanou reprezentací jsme snížili spotřebu malých obvodů popsaných na úrovni tranzistorů použitelných například v technologické knihovně. Dále jsme zavedli novou metodu pro aproximaci aritmetických obvodů, jako jsou sčítačky a násobičky, popsaných na úrovni hradel. S využitím metod formální verifikace navíc celý návrhový proces umožňuje garantovat stanovenou chybu aproximace. Tyto obvody byly využity pro významné snížení příkonu v neuronových sítích pro rozpoznávání obrázků a v diskrétní kosinově transformaci v HEVC kodéru. Pomocí nové chybové metriky nezávislé na rozložení vstupních dat jsme navrhli komplexní aproximativní mediánové filtry vhodné pro zpracování signálů. Disertační práce reprezentuje ucelenou metodiku pro návrh aproximativních obvodů na různých úrovních popisu, která navíc garantuje nepřekročení zadané chyby aproximace.
|
|
|
Implementation of wavelet transform in C++
Valouch, Lukáš ; Hasmanda, Martin (referee) ; Beneš, Radek (advisor)
The aim of this thesis is implementation of wavelet transform algorithm for noise reduction. The noise reduction itself is focused on improving informative capabilities of sonographic (ultrasound) images in medicine. For this purpose, thresholding of detailed coefficients on individual levels of multiresolution analysis was used. Common procedures were not used for searching for the most suitable thresholds of those levels. The alternative concept's design is based on fundamental empirical approach, where the individual thresholds are optimised by evolution algorithms. However, with this algorithmic procedure, more problems manifest regarding the objective evaluation of the success of noise reduction. Because of this, the program uses commonly used parameters such as mean square error of the whole image, linear slope edge approximation, relative contrast of two differently bright and distinct points and the standard deviation of compact surface. Described theoretical knowledge is used in developed application DTWT. It executes multilevel decomposition and reversed reconstruction by discrete time wavelet transform, thresholding of detailed coefficients and final evaluation of performed noise reduction. The developed tool can be used separately to reduce noise. For our purposes, it has been modified in way, that it executed through the component for evolutionary optimization of parameters (Optimize Parameters) in created scenario in RapidMiner program. In the optimization process, this component used evaluation received from DTWT program as fitness function. Optimal thresholds were sought separately for three wavelet families - Daubeschies, Symmlets and Coiflets. The evolution algorithm chose soft threshold for all three wavelet families. In comparison to hard threshold, it is more suitable for noise reduction, but it has tendencies to blur the edges more. The devised method had in most cases greater evaluated success of noise reduction with wavelet transform with threshold search done by evolution algorithms, than commonly used filters. In visual comparison however the wavelet transform introduced some minor depreciating artefacts into the image. It is always about compromise between noise reduction and maximal preservation of image information. Objectively evaluating this dilemma is not easy and is always dependant on subjective viewpoint which in case of sonographic images is that of the attending physician.
|
|
|
Competitive Coevolution in Cartesian Genetic Programming
Skřivánková, Barbora ; Petrlík, Jiří (referee) ; Drahošová, Michaela (advisor)
Symbolic regression is a function formula search approach dealing with isolated points of the function in plane or space. In this thesis, the symbolic regression is performed by Cartesian Genetic Programming and Competitive Coevolution. This task has already been resolved by Cartesian Genetic Programming using Coevolution of Fitness Predictors. This thesis is concerned with comparison of Coevolution of Fitness Predictors with simpler Competitive Coevolution approach in terms of approach effort. Symbolic regression has been tested on five functions with different complexity. It has been shown, that Competitive Coevolution accelerates the symbolic regression task on plainer functions in comparison with Coevolution of Fitness Predictors. However, Competitive Coevolution is not able to solve more complex functions in which Coevolution of Fitness Predictors succeeded.
|
|
|
Evolutionary Algorithms for Neural Networks Learning
Vosol, David ; Rozman, Jaroslav (referee) ; Zbořil, František (advisor)
Main point of this thesis is to find and compare posibilities of cooperation between evolutionary algorithms and neural network learning and their comparison with classical learning technique called backpropagation. This comparison is demonstrated with deep feed-forward neural network which is used for classification tasks. The process of optimalization is via search of optimal values of weights and biases within neural network with fixed topology. We chose three evolutionary approaches. Genetic algorithm, differential evolution and particle swarm optimization algorithm. These three approaches are also compared between each other. The demonstrating program is implemented in Python3 programming language without usage of any third parties libraries focused on deep learning.
|
|
|
Coevolution of Image Filters and Fitness Predictors
Trefilík, Jakub ; Hrbáček, Radek (referee) ; Drahošová, Michaela (advisor)
This thesis deals with employing coevolutionary principles to the image filter design. Evolutionary algorithms are very advisable method for image filter design. Using coevolution, we can add the processes, which can accelerate the convergence by interactions of candidate filters population with population of fitness predictors. Fitness predictor is a small subset of the training set and it is used to approximate the fitness of the candidate solutions. In this thesis, indirect encoding is used for predictors evolution. This encoding represents a mathematical expression, which selects training vectors for candidate filters fitness prediction. This approach was experimentally evaluated in the task of image filters for various intensity of random impulse and salt and pepper noise design and the design of the edge detectors. It was shown, that this approach leads to adapting the number of target objective vectors for a particular task, which leads to computational complexity reduction.
|
|
|
Evolutionary Design of Combinational Circuits on Computer Cluster
Pánek, Richard ; Zachariášová, Marcela (referee) ; Hrbáček, Radek (advisor)
This master's thesis deals with evolutionary algorithms and how them to use to design of combinational circuits. Genetic programming especially CGP is the most applicable to use for this type of task. Furthermore, it deals with computation on computer cluster and the use of evolutionary algorithms on them. For this computation is the most suited island models with CGP. Then a new way of recombination in CGP is designed to improve them. This design is implemented and tested on the computer cluster.
|
|
|
Evolutionary algorithms
Haupt, Daniel ; Polách, Petr (referee) ; Honzík, Petr (advisor)
The first part of this work deals with the optimization and evolutionary algorithms which are used as a tool to solve complex optimization problems. The discussed algorithms are Differential Evolution, Genetic Algorithm, Simulated Annealing and deterministic non-evolutionary algorithm Taboo Search.. Consequently the discussion is held on the issue of testing the optimization algorithms through the use of the test function gallery and comparison solution all algorithms on Travelling salesman problem. In the second part of this work all above mentioned optimization algorithms are tested on 11 test functions and on three models of placement cities in Travelling salesman problem. Firstly, the experiments are carried out with unlimited number of accesses to the fitness function and secondly with limited number of accesses to the fitness function. All the data are processed statistically and graphically.
|
|
|
Aplikace evolučního algoritmu na optimalizační úlohu vibračního generátoru
Nguyen, Manh Thanh ; Kovář, Jiří (referee) ; Hadaš, Zdeněk (advisor)
This thesis will deal with the use of artificial intelligence methods for solving optimization problems with multiple variables. A theorethical part presents problems of global optimization and overview of solution methods. For practical reasons, special attention is paid to evolutionary algorithms. The subject of optimization itself is energy harvester based on a piezoelectric effect. Its nature and modeling is devoted to one chapter. A part of the thesis is the implementation of the SOMA algorithm for finding the optimal parameters of the generator for maximum performance.
|
guest ::
