|
|
Preparation and characterization of fluoride conversion coating on magnesium material
Šmíd, Jakub ; Doskočil, Leoš (referee) ; Březina, Matěj (advisor)
This thesis deals with the preparation of fluoride conversion coating on magnesium material. The effect of the coating time on the formation of the coating was investigated and the corrosion resistance of the resulting coatings was measured. The theoretical part is focused on magnesium and its alloys, powder metallurgy, corrosion and protection against it. Furthermore, conversion coatings, especially fluoride coatings, and their applications are discussed. The review section focuses on the current state of research in the field of fluoride conversion coatings. The experimental part describes the procedures of sample preparation and coating formation. The effect of time on the formation of the coating is investigated by monitoring the microstructure, mainly by elemental analysis of the surface of the samples in the section. Furthermore, the corrosion resistance of the samples is measured by potentiodynamic polarization in NaCl solution.
|
|
|
Agent optimization by means of genetic programming
Šmíd, Jakub ; Neruda, Roman (advisor) ; Kazík, Ondřej (referee)
This thesis deals with a problem of choosing the most suitable agent for a new data mining task not yet seen by the agents. The metric is proposed on the data mining tasks space, and based on this metric similar tasks are identified. This set is advanced as an input to a program evolved by means of genetic programming. The program estimates agents performance on the new task from both the time and error point of view. A JADE agent is implemented which provides an interface allowing other agents to obtain estimation results in real time.
|
|
|
Computational Intelligence Methods in Metalearning
Šmíd, Jakub ; Neruda, Roman (advisor) ; Vanschoren, Joaquin (referee) ; Vomlelová, Marta (referee)
This thesis focuses on the algorithm selection problem, in which the goal is to recommend machine learning algorithms to a new dataset. The idea behind solving this issue is that algorithm performs similarly on similar datasets. The usual approach is to base the similarity measure on the fixed vector of metafeatures extracted out of each dataset. However, as the number of attributes among datasets varies, we may be loosing important information. Herein, we propose a family of algorithms able to handle even the non-propositional representations of datasets. Our methods use the idea of attribute assignment that builds the distance measure between datasets as a sum of distance given by the optimal assignment and an attribute distance measure. Furthermore, we prove that under certain conditions, we can guarantee the resulting dataset distance to be a metric. We carry out a series of metalearning experiments on the data extracted from the OpenML repository. We build up attribute distance using Genetic Algorithms, Genetic Programming and several regularization techniques such as multi-objectivization, coevolution, and bootstrapping. The experiment indicates that the resulting dataset distance can be successfully applied on the algorithm selection problem. Although we use the proposed distance measures exclusively...
|
|
|
Traffic Simulation
Šmíd, Jakub ; Nečaský, Martin (advisor) ; Knap, Tomáš (referee)
Title: Traffic simulation Author: Jakub Šmíd Department: Department of Software Engineering Supervisor: Mgr. Martin Nečaský, Ph. D. Supervisor's e-mail address: necasky@ksi.mff.cuni.c Abstract: The goal of this bachelor thesis is to create a program that simulates the traffic in user defined city. Application mainly watches the status of roads, monitors number of vehicles on them and when the road becomes full, program reports traffic jam and diverts part of the traffic in order to reduce the jam or even completely eliminate it. Besides personal traffic it minitors vehicles of city public transport planned by time table. After the end of simulation it suggest changes that can optimilize time of arrivals of scheduled traffic nearer to scheduled time of arrivals. Keywords: traffic, simulation, jam
|
|
|
Modular and ontogenetic evolution of virtual organisms
Leibl, Marek ; Mráz, František (advisor) ; Šmíd, Jakub (referee)
Increase of computational power and development of new methods in artificial intelligence allow these days many real-world problems to be solved automatically by a~computer program without human interaction. This includes automatized design of walking robots in a~physical virtual environment that can eventually result in construction of real robots. This work compares two different approaches to evolve virtual robotic organisms: artificial ontogeny, where the organism first grows using an~artificial ontogenetic process, and more direct methods. Furthermore, it proposes a~novel approach to evolve virtual robotic organisms: Hypercube-based artificial ontogeny that is combination of artificial ontogeny and Hypercube-based neuroevolution of augmenting topologies (HyperNEAT). Powered by TCPDF (www.tcpdf.org)
|
|
|
Computational Intelligence Methods in Metalearning
Šmíd, Jakub ; Neruda, Roman (advisor) ; Vanschoren, Joaquin (referee) ; Vomlelová, Marta (referee)
This thesis focuses on the algorithm selection problem, in which the goal is to recommend machine learning algorithms to a new dataset. The idea behind solving this issue is that algorithm performs similarly on similar datasets. The usual approach is to base the similarity measure on the fixed vector of metafeatures extracted out of each dataset. However, as the number of attributes among datasets varies, we may be loosing important information. Herein, we propose a family of algorithms able to handle even the non-propositional representations of datasets. Our methods use the idea of attribute assignment that builds the distance measure between datasets as a sum of distance given by the optimal assignment and an attribute distance measure. Furthermore, we prove that under certain conditions, we can guarantee the resulting dataset distance to be a metric. We carry out a series of metalearning experiments on the data extracted from the OpenML repository. We build up attribute distance using Genetic Algorithms, Genetic Programming and several regularization techniques such as multi-objectivization, coevolution, and bootstrapping. The experiment indicates that the resulting dataset distance can be successfully applied on the algorithm selection problem. Although we use the proposed distance measures exclusively...
|
|
|
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)
|
|
|
Study of the composition and production of household waste in the Czech Republic
Šmíd, Jakub ; Benešová, Libuše (advisor) ; Kotoulová, Zdenka (referee)
In many countries they set targets formunicipal waste reuse and for reduction of the amount landfilled. To develop effective strategies requires knowledge of reliable information on the composition of municipal waste. At present, however, in most European countries and the rest of theworld use different methods for the analysis of the composition of waste, which vary considerably, not only in scale but also a focus. As a part of this work has been carried out research on total waste production in households with anemphasis on organic waste, using new research methods. It was determined during production regard less of any subsequent disposal method. This research was supplemented by questionnaire surfy focused on household waste management. There search results showed that the average proportion of individual components in household waste is 25% for paper, 7% for plastic, 12%for glass, 2% for metal, 28% for organic waste, 10% for animal waste, another 16% of waste and 1% for.hazardous waste . It was also not found that the production of individual components of household waste differed significantly between various types of dwelling. They have not been demonstrated a signifiant relationship between waste production and number of members in the household. There search results show that, if...
|
|
|
Creating bidding strategies for double auction markets using the genetic programming
Hudec, Tobiáš ; Pilát, Martin (advisor) ; Šmíd, Jakub (referee)
Computational economics investigates economic phenomena by computer simulations. For these simulations it needs computer programs which can simu- late behaviour of people. It is possible to use artificial intelligence to simulate behaviour of people. We investigate ways of using a type of artificial intelligence called genetic programming in a trading system called double auction. We de- vise a modification of genetic programming in order to improve its performance in double auction. We experimentally verify whether trading controlled by ge- netic programming has properties that we would expect from trading conducted by people. It turns out that genetic programming approximates behaviour of pe- ople well, and is therefore a feasible tool for computational economics. 1
|
|
|
Modular and ontogenetic evolution of virtual organisms
Leibl, Marek ; Mráz, František (advisor) ; Šmíd, Jakub (referee)
Increase of computational power and development of new methods in artificial intelligence allow these days many real-world problems to be solved automatically by a~computer program without human interaction. This includes automatized design of walking robots in a~physical virtual environment that can eventually result in construction of real robots. This work compares two different approaches to evolve virtual robotic organisms: artificial ontogeny, where the organism first grows using an~artificial ontogenetic process, and more direct methods. Furthermore, it proposes a~novel approach to evolve virtual robotic organisms: Hypercube-based artificial ontogeny that is combination of artificial ontogeny and Hypercube-based neuroevolution of augmenting topologies (HyperNEAT). Powered by TCPDF (www.tcpdf.org)
|
guest ::
RSS feed.