|
|
Advanced Evolutionary Image Filtering
Saranová, Ivana ; Drábek, Vladimír (referee) ; Bidlo, Michal (advisor)
This work aims to use cellular automata with a transition function of conditionally matching rules designed by the evolution strategy for the removal of noises of different types and intensities from digital images. The proposed method improves the original concept of conditionally matching rules by modifying the right side of the rule, extending it from a single value to a selection of functions. Furthermore, various evolution strategy setups were explored, including usage of different noise models for evolution, training on partially damaged images, and other setups, resulting in high-quality filters for each noise model. Comparing these filters to the existing methods shows great improvement from the original approach and the ability to evolutionarily design filters that are placed among the top methods quality-wise.
|
|
|
Quantum-Inspired Optimisation Algorithms
Kosík, Dominik ; Sekanina, Lukáš (referee) ; Bidlo, Michal (advisor)
The focus of this work is an implementation of the chosen quantum-inspired optimisation algorithm and its modifications, which will be compared at the end of the work. As the optimisation algorithm was chosen simulated quantum annealing algorithm. The first part of the work will lay the theoretical groundwork of standard optimisation algorithms used in this work, physics from which the inspiration for the simulated quantum annealing originates, and a description of the chosen algorithm. The second part will focus on the implementation of the algorithms on the selected problems. The selected problems are travelling salesman problem, searching rules for cellular automaton and MAX-SAT problem. The last part will contain the proposed modifications of the simulated quantum annealing, a comparison of the basic variant and standard optimisations algorithms, and an evaluation of the results.
|
|
|
Evolutionary Design of Quantum Operator
Kraus, Pavel ; Mrázek, Vojtěch (referee) ; Bidlo, Michal (advisor)
The goal of this thesis is to utilize various evolutionary algorithms for quantum operator design in the form of unitary matrices in direct representation. Evolution strategy, differential evolution, Particle Swarm Optimization and artificial bee colony algorithms were chosen. In this thesis, the third and fourth algorithms were used for the first time in relation to quantum operator design. The experiments have shown that the utilization of direct representation gives results of acceptable quality.
|
|
|
Evolutionary Optimization of Freight Transportation
Beránek, Michal ; Drahošová, Michaela (referee) ; Bidlo, Michal (advisor)
The following thesis deals with optimization of freight transport planning. The goal is to minimize expenses connected to transportation, which emerge from travelled distance. The expenses can be heavily reduced, if the routes are correctly planned, especially when there is a large number of customers to be served. This thesis focuses on solving the problem by using the evolutional algorithms, that are optimization methods based on principles of evolution. Thesis concentrates on Heterogeneous Fixed Fleet Vehicle Routing Problem. Thesis introduces multiple evolutional algorithms and their results are compared. The best algorithm, evolutional strategy with local neighbourhood search, achieves similar, for certain tasks even better results, than other existing evolutional algorithms, created to solve given problem.
|
|
|
Evolutionary approaches to image representation and generation
Romanský, Patrik ; Neruda, Roman (advisor) ; Vidnerová, Petra (referee)
This thesis focuses on exploring different variants of evolutionary algorithms in the area of image data representation and generation. In the contrast of the majority of similar works, this work differs in modular approach to the creation of evolutionary algorithms. The aim of this work is to create an extensible library for creating evolutionary algorithms and comparing the algorithms based on real image data. Compared types of evolutionary algorithms are genetic algorithm, CMA-ES and Differential evolution. Based on experiments, we assessed the success rate of individual evolutionary algorithms and proposed a parallelization of the CMA-ES method.
|
|
|
Application of Evolutionary Algorithms in Quantum Computing
Žufan, Petr ; Mrázek, Vojtěch (referee) ; Bidlo, Michal (advisor)
In this thesis, an evolutionary system for searching quantum operators in the form of unitary matrices is implemented. The aim is to propose several representations of candidate solutions and settings of the evolutionary algorithm. Two evolutionary algorithms were applied: the genetic algorithm and evolutionary strategy. Furthermore, a method of generating a unitary matrix is presented which is used for the first time for this task. This method is in some aspects better than the previous ones. Finally, a comparison of all used techniques is shown in experiments.
|
|
|
Evolutionary Prediction of Time Series
Křivánek, Jan ; Bidlo, Michal (referee) ; Sekanina, Lukáš (advisor)
This thesis summarizes knowledge in the field of time series theory, method for time series analysis and applications in financial modeling. It also resumes the area of evolutionary algorithms, their classification and applications. The core of this work combines these knowledges in order to build a system utilizing evolutionary algorithms for financial time series forecasting models optimization. Various software engineering techniques were used during the implementation phase (ACI - autonomous continual integration, autonomous quality control etc.) to ensure easy maintainability and extendibility of project by more developers.
|
|
|
Evolutionary Design of Boolean Functions for Cryptography
Dvořák, Jan ; Vašíček, Zdeněk (referee) ; Husa, Jakub (advisor)
The goal of this bachelor's thesis is to compare various selection methods used in cartesian genetic programming applied to a problem of various types of cryptographically significant boolean functions. I focused on these selection methods: evolutionary strategies (1+lambda) and (1,lambda), tournament selection and roulette selection. The chosen problem was solved by an implementation of CGP with the above-mentioned selection methods and by a statistical evaluation of data acquired from conducted experiments. Evaluation of mentioned data has shown that the best results in case of bent functions were achieved while using (1+lambda) evolutionary strategy. The roulette selection performed the best in case of balanced functions with high nonlinearity.
|
|
|
Evolutionary Algorithms for the Control of Heterogeneous Robotic Swarms
Karella, Tomáš ; Pilát, Martin (advisor) ; Balcar, Štěpán (referee)
Robotic swarms are often used for solving different tasks. Many articles are focused on generating robot controllers for swarm behaviour using evolutionary algorithms. Most of them are nevertheless considering only homogenous robots. The goal of this thesis is to use evolutionary algorithms for behaviours of heterogeneous robotic swarms. A 2D simulation was implemented to explore swarm controller optimization methods with the ability to create custom scenarios for robotic swarms. We tested differential evolution and evolution strategies on three different scenarios.
|
|
|
Neuroevolution Principles and Applications
Herec, Jan ; Strnadel, Josef (referee) ; Bidlo, Michal (advisor)
The theoretical part of this work deals with evolutionary algorithms (EA), neural networks (NN) and their synthesis in the form of neuroevolution. From a practical point of view, the aim of the work is to show the application of neuroevolution on two different tasks. The first task is the evolutionary design of the convolutional neural network (CNN) architecture that would be able to classify handwritten digits (from the MNIST dataset) with a high accurancy. The second task is the evolutionary optimization of neurocontroller for a simulated Falcon 9 rocket landing. Both tasks are computationally demanding and therefore have been solved on a supercomputer. As a part of the first task, it was possible to design such architectures which, when properly trained, achieve an accuracy of 99.49%. It turned out that it is possible to automate the design of high-quality architectures with the use of neuroevolution. Within the second task, the neuro-controller weights have been optimized so that, for defined initial conditions, the model of the Falcon booster can successfully land. Neuroevolution succeeded in both tasks.
|
guest ::
