|
|
Gut microbiome and its changes related to therapy of chronic diseases
Hurych, Jakub ; Cinek, Ondřej (advisor) ; Kolář, Milan (referee) ; Falt, Přemysl (referee)
This work examines the response of the gut microbiome to therapeutic interventions in three longitudinal studies of chronic gastrointestinal diseases: Crohn's disease, celiac autoimmunity and irritable bowel syndrome. Multiple methods of stool microbiome analysis (especially massively parallel 16S rDNA or 18S rDNA sequencing and metagenomic sequencing) followed by bioinformatic and statistical analysis were used. In Crohn's disease, we detected a previously undescribed secondary nature of changes in the gut bacteriome after anti-TNF treatment. In celiac disease autoimmunity, where previous works described an effect of probiotic intervention on serological markers of the disease, the gut bacteriome and metabolome, we described the absence of significant changes in beneficial gut protozoa. In irritable bowel syndrome, we observed a significant response of the bacteriome after administering four doses of mixed microbiota transplantation but no response in the reduction of clinical symptoms. The results of these studies could contribute to a better understanding of the gut microbiome's role in the pathogenesis of these serious diseases. Keywords: microbiome, Crohn's disease, celiac disease, irritable bowel syndrome
|
|
|
Playing Games Using Neural Networks
Buchal, Petr ; Kolář, Martin (referee) ; Hradiš, Michal (advisor)
The aim of this bachelor thesis is to teach a neural network solving classic control theory problems and playing the turn-based game 2048 and several Atari games. It is about the process of the reinforcement learning. I used the Deep Q-learning reinforcement learning algorithm which uses a neural networks. In order to improve a learning efficiency, I enriched the algorithm with several improvements. The enhancements include the addition of a target network, DDQN, dueling neural network architecture and priority experience replay memory. The experiments with classic control theory problems found out that the learning efficiency is most increased by adding a target network. In the game environments, the Deep Q-learning has achieved several times better results than a random player. The results and their analysis can be used for an insight to reinforcement learning algorithms using neural networks and to improve the used techniques.
|
|
|
E-commerce Design
Kolář, Michal ; Dyk, Tomáš (referee) ; Luhan, Jan (advisor)
Bachelor’s thesis deals with a creation of design and implementation of e-commerce for the newly formed company, focused primarily on Apple devices service. The work focuses on the analysis of the company, requirements and implementation of e-commerce with innovatice elements.
|
|
| |
|
|
Support for Organization of Festival Contests
Pavelek, Miroslav ; Kolář, Martin (referee) ; Křivka, Zbyněk (advisor)
This bachelor thesis is focused on making supporting information system for organization of contests on Animefest festival. It is a web application based on framework Django written in Python language. The system will replace the current web solution and will have new functionalities. For system administrator it provides the creationg and managing new contests and also for managing entries. Contestants will have the possibility to register to the contest and manage their entries.
|
|
|
Deep Neural Networks for Reinforcement Learning in Real-Time Strategy
Barilla, Marco ; Dobeš, Petr (referee) ; Kolář, Martin (advisor)
Machine learning is one of the fastest growing branches of modern science. It is a subfield of artificial intelligence research that is interested the problem of making computers help us solve complex modern problems. Games play an important role in this field because they represent the perfect environment for testing of new approaches and benchmarking against human performance. Starcraft 2 is currently in the spotlight, thanks to its broad playerbase and its complexity. The practical goal of this paper is to create an advantage actor critic agent that is able to operate in the environment of this game.
|
|
|
Administration of development documetation over WWW I
Kolář, Miroslav ; Sysel, Petr (referee) ; Balík, Miroslav (advisor)
Master`s thesis is based on the assignment of company Honeywell to create integral system for saving and management of development documentation. Designed system will introduce data warehouse with transparent display of documents being connected with development of customer’s requirements, products and tests, test runs. Solution consists of two parts: A – Requirements and Products and B – Tests, Test runs and Person, whereas Master`s thesis deals with the part B. For system creation utilities of following programming languages were used HTML, PHP, JavaScript and database system MySQL. In the first phase was designed database structure (tables and relations between them). Except that was designed structure for authorization algorithm and integral concept of whole application was created. The second phase is realization. At first are created supporting algorithms. It is servicing database by functions, uniform displaying results from database and authorized access of users to the application. Further data library for integral display data forms, based on used template system Smarty and other libraries, f.e. attribution of the files to records in SQL tables. Important part was implementation of designed database system and related programs. In Tests and Test runs there is user interface established and individual relations between them. According to assignment all fundamental programs were designed together with support instruments. Individual source files are saved in transparent structure of folders, so this structure is possible to apply on any computer, where are installed programs to web hosting. The result of Master`s thesis is then integrated progress report together with functional source codes.
|
|
|
Design of Forming Tool and Machinery for Couting out of Pad
Kolář, Milan ; Štroner, Marek (referee) ; Dvořák, Milan (advisor)
The Diploma Thesis presents a draft of designated automatic apparatus for cutting out and assembling pads for pressure control valve (DRV2). The theoretical chapter of the Thesis contains selected findings from theory and technology for material cutting. The next chapter contains design solutions for a designated apparatus for cutting and assembling pressure control valves. The pads are etched from 0,08 mm steel plates X2CrNi Mol7-12-2. The shape of the pads is etched on a 35mm-wide belt that is wound up onto a drum. The draft of this new technological facility has been elaborated in accordance with the most suitable technological variant. The cutting tool is made of steel 19 437 that is thermally processed at HRC 62-4. Estimated annual volume of production series is 2 mil. pcs. The cutting and assembly processes are executed with the aid of this designated automatic apparatus.
|
|
|
Deep Learning for Image Recognition
Munzar, Milan ; Kolář, Martin (referee) ; Hradiš, Michal (advisor)
Neural networks are one of the state-of-the-art models for machine learning today. One may found them in autonomous robot systems, object and speech recognition, prediction and many others AI tasks. The thesis describes this model and its extension which is used in an object recognition. Then explains an application of a convolutional neural networks(CNNs) in an image recognition on Caltech101 and Cifar10 datasets. Using this exemplar application, the thesis discusses and measures efficiency of techniques used in CNNs. Results show that the convolutional networks without advanced extensions are able to reach a 80\% recognition accuracy on Cifar-10 and a 37\% accuracy on Caltech101.
|
|
|
Generating training data with neural networks
Ševčík, Pavel ; Kolář, Martin (referee) ; Hradiš, Michal (advisor)
The aim of this thesis was to prepare a training data set for traffic sign detection using generative neural networks. The solution uses a modified U-Net architecture and several experiments with the application of styles using AdaIN layers as in the StyleGAN model have been conducted. By extending the real GTSDB data set with the generated images, mean average precision of 80.36 % has been achieved, which yields an improvement of 19.27 % compared to the mean average precision of the detection model trained on real data only.
|
guest ::
