National Repository of Grey Literature 263 records found  previous11 - 20nextend  jump to record: Search took 0.01 seconds. 
Visualizing Neuroevolution in Neural Network Learning
Bednář, Martin ; Janoušek, Vladimír (referee) ; Zbořil, František (advisor)
This thesis examines options for neural network learning achieved by means of neuroevolution, examines general functioning of neuroevolution, design and implementation of neuroevolution and marginally deals with design and implementation of feed-forward neural networks with fully connected layers. The goal of this thesis is to introduce program, that executes neuroevolutionary algorithm and separate graphic application, which encapsulates this program for easier use and for display of graphic output of the program visualizing problem-solving capabilities of neural networks created by neuroevolution. The end part of the thesis is devoted to experiments done on the created program.
Semantic Analysis of Parish Record
Kaňkovský, Adam ; Zbořil, František (referee) ; Rozman, Jaroslav (advisor)
The aim of this work is to design and implement an application for semantic analysis of matrix records, which will take as input text obtained from a matrix scan. The extracted information will then be entered into the appropriate fields of the table.
Support for the Processing of Intentions
Mechl, Marek ; Zbořil, František (referee) ; Zbořil, František (advisor)
The subject of this bachelor's thesis is closely related to the Intention Progression Competition. The primary objective of the competition is to develop an agent that can effectively solve predefined problems. These problems are provided to the agent in XML file format, which includes a detailed description of the environment and tasks in the form of a goal-plan tree, adhering to the structural guidelines stipulated by the competition. The agent's task is to determine the most appropriate intention to pursue next. The aim of this thesis is to design a system for agents that enables them to work with the given problems by parsing the input file and transforming the tree structure into objects. The agent interacts with the system by issuing commands, allowing them to perform actions and achieve their objectives.
Deep Learning Methods for Machine Playing the Scotland Yard Board Game
Hrkľová, Zuzana ; Janoušek, Vladimír (referee) ; Zbořil, František (advisor)
Táto práca sa zaoberá metódami hlbokého učenia, ktoré sú aplikovateľné na stolné hry s neurčitosťou. V rámci práce boli naštudované princípy učenia s posilňovaním, s hlavným zameraním na Q-learning algoritmy, spomedzi ktorých bol vybraný Deep Q-Network algoritmus. Ten bol následne implementovaný na zjednodušených pravidlách stolnej hry Scotland Yard. Konečná implementácia bola porovnaná s metódami Alpha-Beta a Monte Carlo Tree Search. S výsledkov vyplinulo, že schovávaný hráč riadený DQN algoritmom predstavoval pre ostatné metódy najťažšieho protihráča, narozdiel od hľadajúcich hráčov, ktorým sa nepodarilo zlepšiť existujúce riešenia. Napriek tomu, že implementovaná metóda nedosiahla lepšie výsledky oproti doposiaľ existujúcim metódam, ukázalo sa, že potrebuje najmenej výpočetných zdrojov a času na vykonanie daného ťahu. To ju robí najperspektívnejšou zo spomínaných metód na budúcu možnú implementáciu originálnej verzie danej hry.
Resolution Method Visualisation
Kasem, Jakub ; Zbořil, František (referee) ; Rozman, Jaroslav (advisor)
This bachelor thesis is focused on the topic of automated reasoning using the resolution method. The theoretical part of the thesis focuses on predicate logic, its operations and laws, the resolution method and its processes, state space search by uninformed algorithms, and tools for automated reasoning programming. The thesis further describes the implementation of a educational application visualizing the resolution method. The application automatically proves the truth value of a given statement by finding a contradiction between the negation of the statement and the set of premises and reveals the solution procedure. In the conclusion of the thesis, the implementation of the solution and the content of the work are evaluated and possible improvements are suggested.
Conversion of Models between Machine Learning Frameworks for Mobile Platforms
Pavella, Martin ; Zbořil, František (referee) ; Kočí, Radek (advisor)
Machine learning frameworks use various formats to represent and store models of deep neural networks (DNN). One of the most commonly used ones is Open Neural Network Exchange (ONNX). Developing drivers for hardware accelerators on embedded systems is expensive, and ONNX is rarely supported. The necessary software support is typically only implemented for the TensorFlow Lite (TFLite) DNN model format. Currently, the options for conversion of pre-trained ONNX models to TFLite are inadequate and produce suboptimal models. This work focuses on designing and developing a direct converter of ONNX models to TFLite, which produces as optimal models as possible. The resulting program was verified on real models in collaboration with the NXP company. The models produce identical outputs after conversion and their inference speed on target platforms is significantly higher.
Systém pro doporučování filmů
Hnatovskyj, Vítek ; Rozman, Jaroslav (referee) ; Zbořil, František (advisor)
This work focuses on a movie recommendation system. First, the issue of recommendation systems in general is described, and the various types of these systems are outlined. The main goal is to implement a system that recommends relevant movies to the user based on their preferences. This system is hybrid, consisting of a content-based system and a collaborative filtering system. To test the system, a simple application is implemented, which is also described in this work. Subsequently, the system is evaluated using offline metrics and user testing.
Modeling System Based on Archival Data
Haškovec, Martin ; Kočí, Radek (referee) ; Zbořil, František (advisor)
The aim of this work is to design and implement scripts in PHP for converting output archived data from Prolog to the MySQL web application database. The application script should be used to transfer the writing of archived data processed by the Prolog language to others relational databases used for data archiving and their further presentation.
Support for Searching in Parish Books
Konetzný, Jakub ; Zbořil, František (referee) ; Rozman, Jaroslav (advisor)
The aim of this work is to design and implement a web application for support for searching in parish book. The created application allows the user to upload his GEDCOM file from the missing events in this file to automatically create records to which he assigns parish book where the event could be located. The created solution allows you to write notes for individual records and parish book.
Playing the Board Game Stratego by Computer
Irovský, Dominik ; Šátek, Václav (referee) ; Zbořil, František (advisor)
The topic of this thesis is the board game of Stratego. This game features incomplete information. The goal of this thesis is research of existing game playing algorithms and, design and implementation of new solution. For the new solution modified version of Monte Carlo Tree Search was used. The solution was implemented as a console application with possibility of future expansion. Functionality of the solution was validated and tested using experiments. Effectivity of the final algorithm was satisfying

National Repository of Grey Literature : 263 records found   previous11 - 20nextend  jump to record:
Interested in being notified about new results for this query?
Subscribe to the RSS feed.