|
|
6 who loses! game - AI server
Hruška, Michal ; Holan, Tomáš (advisor) ; Peška, Ladislav (referee)
This bachelor thesis deals with the implementation of a web server for playing the card game 6 who loses! The game can be played against other people over the internet or against a computer. Part of the thesis is also the design of a language for the creation of an artificial intelligence. Using the designed language, everybody can create their own algorithm. The algorithm can be used in the game or tested by a simulation, which is part of the implementation. The server includes an artificial intelligence created in this language by using the evolutionary algorithm. Powered by TCPDF (www.tcpdf.org)
|
|
|
Converged Networks and Traffic Tomography by Using Evolutionary Algorithms
Oujezský, Václav ; Sýkora, Jiří (referee) ; Polívka, Michal (referee) ; Škorpil, Vladislav (advisor)
Nowadays, the traffic tomography represents an integral component in converged networks and systems for detecting their behavioral characteristics. The dissertation deals with research of its implementation with the use of evolutionary algorithms. The research was mainly focused on innovation and solving behavioral detection data flows in networks and network anomalies using tomography and evolutionary algorithms. Within the dissertation has been proposed a new algorithm, emerging from the basics of the statistical method survival analysis, combined with a genetics’ algorithm. The proposed algorithm was tested in a model of a self-created network probe using the Python programming language and Cisco laboratory network devices. Performed tests have shown the basic functionality of the proposed solution.
|
|
|
Utilizing artificial neural networks to accelerate evolutionary algorithms
Wimberský, Antonín ; Holeňa, Martin (advisor) ; Gemrot, Jakub (referee)
In the present work, we study possibilities of using artificial neural networks for accelerating of evolutionary algorithms. Improving consists in decreasing in number of calls to the fitness function, the evaluation of which is in some kinds of optimization problems very time- consuming and expensive. We use neural network as a regression model, which serves for fitness estimation in a run of evolutionary algorithm. Together with the regression model, we work also with the real fitness function, which we use for re-evaluation of individuals that are selecting according to a beforehand chosen strategy. These individuals re-evaluated by the real fitness function are used for improving the regression model. Because a significant number of individuals are evaluated only with the regression model, the number of calls to the real fitness function, that is needed for finding of a good solution of the optimization problem, is substantially reduced.
|
|
|
Mutation in Cartesian Genetic Programming
Končal, Ondřej ; Hrbáček, Radek (referee) ; Wiglasz, Michal (advisor)
This thesis examines various kinds of mutations in the Cartesian Genetic Programming (CGP) on tasks of symbolic regression. The CGP is kind of evolutionary algorithm which operates with executable structures. Programs in CGP are evolved using mutation, which leads to offspring evaluation, which is the most time-consuming part of the algorithm. Finding more suitable kind of mutation can significantly accelerate the creation of new individuals and thus, reduce the time necessary to find a satisfactory solution. This thesis presents four different mutations for CGP. Experiments compare these mutation operators to solve five tasks of symbolic regression. Experiments have shown that a choice of suitable mutation can almost double the computing speed in comparison to the standard mutation.
|
|
|
Co-Learning in Cartesian Genetic Programming
Korgo, Jakub ; Grochol, David (referee) ; Wiglasz, Michal (advisor)
This thesis deals with the integration of co-learning into cartesian genetic programming. The task of symbolic regression was already solved by cartesian genetic programming, but this method is not perfect yet. It is relatively slow and for certain tasks it tends not to find the desired result. However with co-learning we can enhance some of these attributes. In this project we introduce a genotype plasticity, which is based on Baldwins effect. This approach allows us to change the phenotype of an individual while generation is running. Co-learning algorithms were tested on five different symbolic regression tasks. The best enhancement delivered in experiments by co-learning was that the speed of finding a result was 15 times faster compared to the algorithm without co-learning.
|
|
|
Neural Networks Classifier Design using Genetic Algorithm
Tomášek, Michal ; Vašíček, Zdeněk (referee) ; Mrázek, Vojtěch (advisor)
The aim of this work is the genetic design of neural networks, which are able to classify within various classification tasks. In order to create these neural networks, algorithm called NeuroEvolution of Augmenting Topologies (also known as NEAT) is used. Also the idea of preprocessing, which is included in implemented result, is proposed. The goal of preprocessing is to reduce the computational requirements for processing of benchmark datasets for classification accuracy. The result of this work is a set of experiments conducted over a data set for cancer cells detection and a database of handwritten digits MNIST. Classifiers generated for the cancer cells exhibits over 99 % accuracy and in experiment MNIST reduces computational requirements more than 10 % with bringing negligible error of size 0.17 %.
|
|
|
Toolbox for multi-objective optimization
Marek, Martin ; Hurák,, Zdeněk (referee) ; Kadlec, Petr (advisor)
This paper deals with multi-objective optimization problems (MOOP). It is explained, what solutions in multi-objetive search space are optimal and how are optimal (non-dominated) solutions found in the set of feasible solutions. Afterwards, principles of NSGA-II, MOPSO and GDE3 algorithms are described. In the following chapters, benchmark metrics and problems are introduced. In the last part of this paper, all the three algorithms are compared based on several benchmark metrics.
|
|
| |
|
|
Cellular Automaton in Dynamical Environment
Bendl, Jaroslav ; Jaroš, Jiří (referee) ; Bidlo, Michal (advisor)
This bachelor thesis focuses on the method of evolution of cellular automaton capable of self-repair after being damaged by external environment. The described method is based on cellular programming algorithm and uses principles of biological development. Experiments leading to verification of regenerative ability for cellular automaton evolved by this approach are presented in this work.
|
|
|
Evolutionary Design Using Random Boolean Networks
Mrnuštík, Michal ; Žaloudek, Luděk (referee) ; Bidlo, Michal (advisor)
This master's thesis introduces the Random Boolean Networks as a developmental model in the evolutionary design. The representation of the Random Boolean Networks is described. This representation is combined with an evolutionary algorithm. The genetic operators are described too. The Random Boolean Networks are used as the developmental model for the evolutionary design of the combinational circuits and the sorting networks. Moreover a representation of the Random Boolean Networks for the design of image filters is introduced. The proposed methods are evaluated in different case-studies. The results of the experiments are discussed together with the potential improvements and topics of the next research.
|
guest ::
