|
|
Using Cellular Automata for Data Compression
Polák, Marek ; Trunda, Otakar (advisor) ; Mráz, František (referee)
In this thesis we research the possibilities of using cellular automata for lossless data compression. We describe the classification of cellular automata and their current usage. We study the properties of various types of elementary cellular automata (i.e. Wolfram rules), describe their equivalence classes, the ways of forward as well as backward simulation, we examine the rules with interesting behavior. The states provided by these rules are evaluated in terms of their orderliness (e.g. the ratio of living cells or approximation of entropy). We implement some standard compression algorithms and compare them in terms of usability for best rated states. By application of acquired knowledge we propose a new compression algorithm, test it on text and image data and compare the results with traditional compression algorithms. Powered by TCPDF (www.tcpdf.org)
|
|
|
Computational Intelligence for Financial Market Prediction
Řeha, Filip ; Pilát, Martin (advisor) ; Mráz, František (referee)
Financial markets are characterized by uncertainty, which is associated with the future progress of world economics and corporations. The ability of an individual to forecast future market behaviour at least to a certain extent would give him an important competitive advantage on the market. The aim of this work is to explore neural networks and genetic programming as possible tools which could be used for financial markets forecasting and apply them on historical financial data. Experiments using neural networks and genetic programming were performed and the results show, that both tools can be employed successfully. On average, neural networks outperformed genetic programming in our experiments. In order to evaluate and visualize the results of our created strategies, the MarketForecaster application was implemented. Powered by TCPDF (www.tcpdf.org)
|
|
|
Genetic programming in Swift for human-competitive evolution
Mánek, Petr ; Mráz, František (advisor) ; Gemrot, Jakub (referee)
Imitating the process of natural selection, evolutionary algorithms have shown to be efficient search techniques for optimization and machine learning in poorly understood and irregular spaces. In this thesis, we implement a library containing essential implementation of such algorithms in recently unveiled programming language Swift. The result is a lightweight framework compatible with Linux- based computing clusters as well as mobile devices. Such wide range of supported platforms allows for successful application even in situations, where signals from various sensors have to be acquired and processed independently of other devices. In addition, thanks to Swift's minimalistic and functional syntax, the implementation of bundled algorithms and their sample usage clearly demonstrates fundamentals of genetic programming, making the work usable in teaching and quick prototyping of evolutionary algorithms. Powered by TCPDF (www.tcpdf.org)
|
|
|
Artificial Player for Hearthstone Card Game
Ohman, Ľubomír ; Gemrot, Jakub (advisor) ; Mráz, František (referee)
The goal of this work was to create an artificial agent that is able to learn how to play Hearthstone with given deck of cards. We decided to use Q-learning algorithm to achieve it. The side effect of this work is the transformation of the simple simulator of Hearthstone into the framework for developing Artificial Intelligence in this game. For the purpose of training and evaluation, commonly played strategies served us as inspiration for the testing agents that we developed. This work contains comparison of the table representation of Q-function and the neural network approximation of it. The original goal was fulfilled partially. We were successful in the creation of the learning agent but he is only able to learn one specific strategy.
|
|
|
Efficient video retrieval using complex sketches and exploration based on semantic descriptors
Blažek, Adam ; Lokoč, Jakub (advisor) ; Mráz, František (referee)
This thesis focuses on novel video retrieval scenarios. More particularly, we aim at the Known-item Search scenario wherein users search for a short video segment known either visually or by a textual description. The scenario assumes that there is no ideal query example available. Our former known- item search tool relying on color feature signatures is extended with major enhancements. Namely, we introduce a multi-modal sketching tool, the exploration of video content with semantic descriptors derived from deep convolutional networks, new browsing/visualization methods and two orthogonal approaches for textual search. The proposed approaches are embodied in our video retrieval tool Enhanced Sketch-based Video Browser (ESBVB). To evaluate ESBVB performance, we participated in international competitions comparing our tool with the state-of-the-art approaches. Repeatedly, our tool outperformed the other methods. Furthermore, we show in our user study that even novice users are able to effectively employ ESBVB capabilities to search and browse known video clips. Powered by TCPDF (www.tcpdf.org)
|
|
|
Maximizing Computational Power by Neuroevolution
Matzner, Filip ; Mráz, František (advisor) ; Pilát, Martin (referee)
Echo state networks represent a special type of recurrent neural networks. Recent papers stated that the echo state networks maximize their computational performance on the transition between order and chaos, the so-called edge of chaos. This work confirms this statement in a comprehensive set of experiments. Afterwards, the best performing echo state network is compared to a network evolved via neuroevolution. The evolved network outperforms the best echo state network, however, the evolution consumes significant computational resources. By combining the best of both worlds, the simplicity of echo state networks and the performance of evolved networks, a new model called locally connected echo state networks is proposed. The results of this thesis may have an impact on future designs of echo state networks and efficiency of their implementation. Furthermore, the findings may improve the understanding of biological brain tissue. 1
|
|
|
Restricted Restarting Automata
Černo, Peter ; Mráz, František (advisor) ; Kutrib, Martin (referee) ; Průša, Daniel (referee)
Restarting automata were introduced as a model for analysis by reduction which is a linguistically motivated method for checking correctness of a sentence. The thesis studies locally restricted models of restarting automata which (to the contrary of general restarting automata) can modify the input tape based only on a limited context. The investigation of such restricted models is easier than in the case of general restarting automata. Moreover, these models are effectively learnable from positive samples of reductions and their instructions are human readable. Powered by TCPDF (www.tcpdf.org)
|
|
|
Evolution and Learning of Virtual Robots
Krčah, Peter ; Mráz, František (advisor) ; Kvasnička, Vladimír (referee) ; Pilát, Martin (referee)
Title: Evolution and Learning of Virtual Robots Author: RNDr. Peter Krčah Department: Department of Software and Computer Science Education Supervisor: RNDr. František Mráz, CSc. Abstract: Evolutionary robotics uses evolutionary algorithms to automatically design both body and controller of a robot. We describe two contributions to automated design of virtual robotic creatures. First, we introduce a nature-inspired method that allows virtual robots to modify their morphology through lifetime learning. We show that such morphological plasticity makes it possible to evolve robots that can dynamically adjust their morphology to the environment they are placed into. We also show that by reshaping the fitness landscape, learning reduces computation cost required to evolve a robot with a given target fitness even in a single environment. In the second contribution, we show that for certain problems in evolutionary robotics, premature convergence to local optima can be avoided by ignoring the original objective and searching for any novel behaviors instead (a technique known as Novelty Search). Keywords: Evolution of Virtual Creatures, Body-brain Coevolution, Morphological Plasticity, Neural Networks, Learning
|
|
|
Applying genetic algorithms for decision trees induction
Šurín, Lukáš ; Mráz, František (advisor) ; Šmíd, Jakub (referee)
Decision trees are recognized and widely used technique for processing and analyzing data. These trees are designed with typical and generally known inductive techniques (such as ID3, C4.5, C5.0, CART, CHAID, MARS). Predictive power of created trees is not always perfect and they often provide a room for improvement. Induction of trees with difficult criterias is hard and sometime impossible. In this paper we will deal with decision trees, namely their creation. We use the mentioned room for improvement by metaheuristic, genetic algorithms, which is used in all types of optimalization. The work also includes an implementation of a new proposed algorithm in the form of plug-in into Weka environment. A comparison of the proposed method for induction of decision trees with known algorithm C4.5 is an integral part of this thesis. Powered by TCPDF (www.tcpdf.org)
|
|
|
IVA Movement Quality Improvement for the Virtual Environment of Unreal Tournament 2004
Macháč, Bohuslav ; Gemrot, Jakub (advisor) ; Mráz, František (referee)
Title: IVA Movement Quality Improvement for the Virtual Environment of Unreal Tournament 2004 Author: Bohuslav Macháč Department / Institute: Department of Software and Computer Science Education Supervisor of the master thesis: Mgr. Jakub Gemrot, Department of Software and Computer Science Education Abstract: PogamutUT2004 is an extension of the Pogamut platform designed for developing intelligent virtual agents (IVAs) in Unreal Tournament 2004. Navigation of IVAs in Pogamut is handled by a navigation system, which uses a navigation graph as an environment abstraction. Navigation mesh is a new, more advanced abstraction, but the existing navigation system is not capable of using its advantages. We created a new navigation system, which exploits advantages of the navigation mesh and solves several other issues of the old one. We show that the new navigation system improves the quality of navigation. To demonstrate the quality improvement, an evaluation framework was created for the comparison of navigation systems. Systems were compared in terms of total number of significant paths on the map, which the system is able to follow, length of the path and time of the navigation. We selected 18 different maps for thorough evaluation and we performed the basic evaluation on 58 other maps. The new system is more...
|
guest ::
