|
|
Hyperparameter optimization in AutoML systems
Pešková, Klára ; Neruda, Roman (advisor) ; Awad, Mariette (referee) ; Kordik, Pavel (referee)
In the last few years, as processing the data became a part of everyday life in different areas of human activity, the automated machine learning systems that are designed to help with the process of data mining, are on the rise. Various metalearning techniques, including recommendation of the right method to use, or the sequence of steps to take, and to find its optimum hyperparameters configuration, are integrated into these systems to help the researchers with the machine learning tasks. In this thesis, we proposed metalearning algorithms and techniques for hyperparameters optimization, narrowing the intervals of hyperparameters, and recommendations of a machine learning method for a never before seen dataset. We designed two AutoML machine learning systems, where these metalearning techniques are implemented. The extensive set of experiments was proposed to evaluate these algorithms, and the results are presented.
|
|
|
Smart Traffic Intersection
Škopková, Věra ; Barták, Roman (advisor) ; Forst, Libor (referee)
This thesis is concerned with the problem of planning paths for autonomous cars through a smart traffic intersection. In this thesis, we describe existing concepts for solving this problem and discuss the possibilities of approaching intersection problems theoretically. Then, we choose one specific approach and design a declarative model for solving the problem. We use that model to perform a series of theoretical experiments to test the throughput and the quality of intersection paths described by different graphs. After that, we translate theoretical plans to actions for real robots and run it. In these experiments, we measure the degree of robots desynchronization and performance success of the plans based on the collision rate. We also describe how to improve action translation to achieve better performance than that for real robots following the straightforward plans.
|
|
|
Interconnection of Recent Strategic Games with Multi-Agent Frameworks
Válek, Lukáš ; Kočí, Radek (referee) ; Zbořil, František (advisor)
This thesis is focused on design of framework for creation an articial opponents in strategy games. We will analyze different types of strategy games and artificial intelligence systems used in these types of games. Next we will describe problems, which can occur in these systems and why agent-based systems makes better artificial opponents. Next we will use knowledge from this research to design and implement framework, which will act as support for creating an artificial intelligence in strategy games.
|
|
|
Adaptive Matchmaking Algorithms for Computational Multi-Agent Systems
Kazík, Ondřej ; Neruda, Roman (advisor) ; Paprzycki, Marcin (referee) ; Diamantini, Claudia (referee)
The multi-agent systems (MAS) has proven their suitability for implementation of complex software systems. In this work, we have analyzed and designed the data mining MAS by means of role-based organizational model. The organiza- tional model and the model of data mining methods have been formalized in the description logic. By matchmaking which is the main subject of our research, we understand the recommendation of computational agents, i.e. agents encap- sulating some computational method, according their capabilities and previous performances. The matchmaking thus consist of two parts: querying the ontol- ogy model and the meta-learning. Three meta-learning scenarios were tested: optimization in the parameter space, multi-objective optimization of data min- ing processes and method recommendation. A set of experiments in these areas have been performed. 1
|
|
|
Distributed Monte-Carlo Tree Search for Games with Team of Cooperative Agents
Filip, Ondřej ; Lisý, Viliam (advisor) ; Majerech, Vladan (referee)
The aim of this work is design, implementation and experimental evaluation of distributed algorithms for planning actions of a team of cooperative autonomous agents. Particular algorithms require different amount of communication. In the work, the related research on Monte-Carlo tree search algorithm, its parallelization and distributability and algorithms for distributed coordination of autonomous agents. Designed algorithms are tested in the environment of the game of Ms Pac-Man. Quality of the algorithms is tested in dependence on computational time, the amount of communication and the robustness against communication failures. Particular algorithms are compared according to these characteristics. Powered by TCPDF (www.tcpdf.org)
|
|
|
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.
|
|
|
Strategic Game in Multi-Agent System Jason
Vais, Roman ; Zbořil, František (referee) ; Král, Jiří (advisor)
This thesis describes artificial inteligence used in developement of computer games, particularly discusses with theory behing artificial inteligence used in real-time strategy games. It deals with implementation of extensions for one such a game. It analyzes posibylities of use multi-agent systems architecture for purposess of artificial inteligence in games. It describes concept of swarm inteligence as suitable but not used tool for developing not only videogame artificial inteligence. Moreover it attempts to find suitable representation of sensation for software agents and shows the difficulties of this task.
|
|
|
Flight Simulator Based on Jason Multi-Agent System
Punčochářová, Pavlína ; Zbořil, František (referee) ; Samek, Jan (advisor)
The aim of this work is to create a 2D flight simulator using Jason multi-agent system and AgentSpeak language. In this simulator, the aircrafts will be represented by agents and their behavior will be described by plans implemented in AgentSpeak language. Agents will use these plans to demonstrate different types of behavior such as detection and collision avoidance, the taking of a position or persecution. The final application consists of two parts - the simulation module and the user interface. Simulation module is the core of simulator and it is created in the Jason system. The user interface provides the possibility to control simulations and creating user simulation models.
|
|
|
Strategic Game in Multi-Agent System Jason
Leško, Matej ; Samek, Jan (referee) ; Král, Jiří (advisor)
Tato práce se zaobírá problematikou koncepce a vývoje multi-agentní tahové strategické hry. Práce analyzuje teorii těchto her a agentních systémů a výsledky této analýzy následně zohledňuje v návrhu samotné hry. Ta implementuje kromě herních konceptů a ovládání, obecně užívaných, dvě úrovně kooperace umělé inteligence, mírnou a komplexní spolupráci. Agent je v této hře hráčem, ovládajícím různé jednotky. Hra je koncipovaná tak, že je jí možné rozšířit o nové druhy inteligence, případně o nové herní jednotky. Závěrečná část práce se soustředí na srovnání jednotlivých úrovní kooperace, na efektivitu jednotlivých umělých inteligencí a také na zhodnocení efektivity implementace hry. S tímto účelem byla vykonána série automatizovaných testů.
|
|
|
Cooperative Game in Multi-Agent System Jason
Husa, Jakub ; Rozman, Jaroslav (referee) ; Král, Jiří (advisor)
The subject of this thesis is creation of a game based on a multi-agent system. First part of the thesis introduces theory of multi-agent systems, strategic computer games, language AgentSpeak and his extension Jason which were used for implementation. Then design of turn-based strategic game Prison Escape is described, which pitches two asymmetrical teams of intelligent agents with different goals against each other. Each team has three modes of intelligent behavior designed for them, which differ by their complexity and extent of inter-agent cooperation. One team can be controlled by a player, or both teams can be controlled by computer. The game is tested on a set of testing maps. The thesis compares the three designed modes of behavior and evaluates their ability to succeed against varying opponents and thinking speed.
|
guest ::
