|
|
Deep Learning AI in Game Environments
Glós, Kristián ; Bobák, Petr (referee) ; Polášek, Tomáš (advisor)
This thesis is focused on analysing deep learning algorithms and their ability to complete given tasks implemented in game environments created via the Unity game engine. Secondary objective was to research and specify possible use-cases of deep learning during game development. The algorithms used fall into Reinforcement learning, Imitation learning and Neuroevolution, while Reinforcement learning was used throughout the whole game scene development cycle. Analysis and results were collected through training the networks in different game scene states and other factors.
|
|
| |
|
|
Planar Antennas on Electromagnetic Bandgap Substrates
Horák, Jiří ; Dědková, Jarmila (referee) ; Škvor,, Zbyněk (referee) ; Raida, Zbyněk (advisor)
Planar antennas are used in several technical applications. The family of planar antennas contains microstrip antennas, which are very popular due to the low weight, low profile, simple manufacturing and easy mass production. Lower gain and excitation of surface waves are disadvantages of microstrip antennas. The propagation of surface waves can be efficiently suppressed if the conventional substrate is replaced by an electromagnetic bandgap (EBG) substrate. Microstrip antennas on EBG substrates have been presented in an open literature for several years. Nevertheless, no published work is devoted to the design of EBG substrates, which can suppress surface waves at several frequencies those cannot be covered by a single bandgap. In order to reach optimum parameters of designed antennas, selected global optimization methods are applied (genetic algorithms, particle swarm optimization, ant colony optimization).
|
|
|
Travelling Salesman Problem
Kolář, Adam ; Rozman, Jaroslav (referee) ; Zbořil, František (advisor)
The aim of this bachelor's thesis is to design a testing environment for the traveling salesman problem and compare the effectiveness of different approaches to the solution. The first part discussed the possibility of genetic algorithms, depending on the setting of a crossover, mutations and population size. In the second part, there is the same problem using two types of neural networks. The representative of the self-learning net was chosen Kohonen neural network. Hopfield neural network represents a method of minimizing the energy function with fixed coefficients. At both neural networks, there were described possible advantages and disadvantages. In the end, all the findings were interpreted in a global context.
|
|
|
Evolutionary Algorithms
Szöllösi, Tomáš ; Mézl, Martin (referee) ; Kozumplík, Jiří (advisor)
The task of this thesis was focused on comparison selected evolutionary algorithms for their success and computing needs. The paper discussed the basic principles and concepts of evolutionary algorithms used for optimization problems. Author programmed selected evolutionary algorithms and subsequently tasted on various test functions with exactly the given input conditions. Finally the algorithms were compared and evaluated the results obtained for different settings.
|
|
|
Flood Prediction in Borovnice - Dalečín Measure Profiles
Hiesböcková, Tereza ; Janál, Petr (referee) ; Starý, Miloš (advisor)
Aim of a work is construction of forecasting models for prediction of flood flows of measuring profile Borovnice – Dalečín on the river Svratka. As a tool for issuing predictions will be used classic hydrological forecasting models, and models based on artificial intelligence methods. Predictive model will be consisting from summer flood flows for the years 1997-2007. In the end of the work will chosen a better method for issuing forecasts
|
|
|
Optimization based on genetic algorithms for image registration
Horáková, Pavla ; Mézl, Martin (referee) ; Harabiš, Vratislav (advisor)
Diploma thesis is focused on global optimization methods and their utilization for medical image registration. The main aim is creation of the genetic algorithm and test its functionality on synthetic data. Besides test functions and test figures algorithm was subjected to real medical images. For this purpose was created graphical user interface with choise of parameters according to actual requirement. After adding an iterative gradient method it became of hybrid genetic algorithm.
|
|
|
Heuristic algorithms in optimization
Šandera, Čeněk ; Popela, Pavel (referee) ; Roupec, Jan (advisor)
Práce se zabývá určením pravděpodobnostních rozdělení pro stochastické programování, při kterém jsou optimální hodnoty účelové funkce extrémní (minimální nebo maximální). Rozdělení se určuje pomocí heuristických metod, konkrétně pomocí genetických algoritmů, kde celá populace aproximuje hledané rozdělení. První kapitoly popisují obecně matematické a stochastické programování a dále jsou popsány různé heuristické metody a s důrazem na genetické algoritmy. Těžiště práce je v naprogramování daného algoritmu a otestování na úlohách lineárních a kvadratických stochastických modelů.
|
|
| |
|
|
Methods and algorithms for face recognition
Soukup, Jiří ; Heriban, Pavel (referee) ; Šťastný, Jiří (advisor)
This work is describing basic methods of face recognition. The methods PCA, LDA, ICA, trace tranfsorm, elastic bunch graph map, genetic algorithm and neural network are described. In practical part, the PCA, PCA + RBF neural network and genetic algorithms are implemented. The RBF neural network is used in the way of clasificator and genetic algorithm is used for RBF NN training in one case and for selecting eigenvectors from PCA method in the other case. This method, PCA + GA, called EPCA, outperform other methods tested in this work on the ORL testing database.
|
guest ::
