|
|
Machine Learning in the Monitoring of Computer Clusters
Adam, Martin ; Pilát, Martin (advisor) ; Balcar, Štěpán (referee)
With the explosion of the number of distributed applications, a new dynamic server environment emerged grouping servers into clusters, whose utilization depends on the cur- rent demand for the application. Detecting and fixing erratic server behavior is paramount for providing maximal service stability and availability. Using standard techniques to de- tect such behavior is yielding sub-optimal results. We have collected a dataset of OS-level performance metrics from a cluster running a streaming distributed application and in- jected artificially created anomalies. We then selected a set of various machine learning algorithms and trained them for anomaly detection on said dataset. We evaluated the algorithms performance and proposed a system for generating notifications of possible erratic behavior, based on the analysis of the best performing algorithm. 1
|
|
|
Collective Robotic Search Using Particle Swarm Optimization
Kunc, Ondřej ; Pilát, Martin (advisor) ; Mráz, František (referee)
With the advancement and miniaturization of electronic devices we experience today a lot of attention is shifting to small robots capable of solving a specific problem collectively. Such problem can be a seeking of a specific target emitting a specific radiation which is detected by the robots depending on its intensity. A technique called particle swarm optimization was proven to be a very useful approach for this task. This technique uses so-called particles moving in a multidimensional space searching for an optimum. This paper deals with the application of this approach to the swarm of robots. A simulator capable of visua- lizing and collecting data was developed and described in this work and practical conclusions were drawn. Modifications and improvements of the algorithm were made to improve the search process in various environmental and robot settings. 1
|
|
|
Terra-Settler Space Real-Time Strategy
Meduna, Lukáš ; Pilát, Martin (advisor) ; Ježek, Pavel (referee)
Terra-Settler is a real-time 3D single player space strategy game. Main focus is on economic simulation, whitch is necessary for achieving game goals. Different scenarios (e.g. conquering planet) are used in game. It is necessary to expand and colonize surrounding planets to achieve victory. API for arti- ficial intelligence is included in the game with sample implementation. This game is multi-platform with support of some mobile platforms. 1
|
|
|
Computational Intelligence for Malware Classification
Tomášek, Jan ; Pilát, Martin (advisor) ; Surynek, Pavel (referee)
As the number of computers and other smart devices grows in every aspect of human life, the amount of malicious software (malware) also grows. Such software tries to disrupt computer usage. Therefore one of the challenges for computer science is to divide the Malware into classes according to its behaviour. The thesis summarizes known ways to look at the problem at hand, some of them are extensions of known approaches, while others are completely new. They are all implemented, tested and compared. We also propose few ideas for future research. Powered by TCPDF (www.tcpdf.org)
|
|
|
Artificial Composition of Multi-Instrumental Polyphonic Music
Samuel, David ; Pilát, Martin (advisor) ; Neruda, Roman (referee)
David Samuel We propose a generative model for artificial composition of both classical and popular music with the goal of producing music as well as humans do. The problem is that music is based on a highly sophisticated hierarchical structure and it is hard to measure its quality automatically. Contrary to other's work, we try to generate a symbolic representation of music with multiple different instruments playing simultaneously to cover a broader musical space. We train three modules based on LSTM networks to generate the music; a lot of effort is put into reducing high complexity of multi-instrumental music representation by a thorough musical analysis. Our work serves mainly as a proof-of-concept for music composition. We believe that the proposed preprocessing techniques and symbolic representation constitute a useful resource for future research in this field. 1
|
|
|
Evolution and Learning of Virtual Robots
Krčah, Peter ; Mráz, František (advisor) ; Kvasnička, Vladimír (referee) ; Pilát, Martin (referee)
Title: Evolution and Learning of Virtual Robots Author: RNDr. Peter Krčah Department: Department of Software and Computer Science Education Supervisor: RNDr. František Mráz, CSc. Abstract: Evolutionary robotics uses evolutionary algorithms to automatically design both body and controller of a robot. We describe two contributions to automated design of virtual robotic creatures. First, we introduce a nature-inspired method that allows virtual robots to modify their morphology through lifetime learning. We show that such morphological plasticity makes it possible to evolve robots that can dynamically adjust their morphology to the environment they are placed into. We also show that by reshaping the fitness landscape, learning reduces computation cost required to evolve a robot with a given target fitness even in a single environment. In the second contribution, we show that for certain problems in evolutionary robotics, premature convergence to local optima can be avoided by ignoring the original objective and searching for any novel behaviors instead (a technique known as Novelty Search). Keywords: Evolution of Virtual Creatures, Body-brain Coevolution, Morphological Plasticity, Neural Networks, Learning
|
|
| |
|
|
Creating the Game Strategies for PuppetWars Using Neuroevolution
Šmelko, Adam ; Pilát, Martin (advisor) ; Matzner, Filip (referee)
In recent years the gaming industry has been on increase. In order to maintain competitiveness gaming companies are required to develop more and more compelling computer games what implies the presence of the very responsive artificial intelligence controlling the game elements, on which our work focuses. We have implemented a simple 2D programming game where we have experimented with the artificial intelligence in it trying to create a strategy beeing able to compete with human. We have explored several variations of learning through the evolutionary strategy applied to neural networks and we have created game characters worthy of being an equal opponent to the game user.
|
|
|
Machine Learning for Simulated Military Vehicles
Závorka, Kamil ; Gemrot, Jakub (advisor) ; Pilát, Martin (referee)
Recent research in the field of neural networks has shown that this is a very promising area of artificial intelligence. Results of the research indicate that neural networks are currently able to at least match humans in many areas. One of the intensively researched sectors is the driving of autonomous vehicles. Although most people focus on autonomous vehicles in the real world, this new artificial intelligence can also be beneficial for driving in the digital world. As more and more activities and experiments are being moved from real environments to simulated environments, the demands on the quality of artificial intelligence found in digital environments are also increasing. The aim of this work was to explore the possibilities of artificial intelligence based on deep feedback learning in the field of parking simulated vehicles. Based on this research, we created a prototype neural network and evaluated this prototype during parking in a simulated environment.
|
|
|
Game Based on Evolutionary Algorithms
Kotěšovcová, Tereza ; Pilát, Martin (advisor) ; Gemrot, Jakub (referee)
The evolution of the movement of virtual creatures is an interesting application of evolutionary algorithms in the field of Artificial Life. The goal of this thesis was to explore the possibilities of its use in computer games. As part of this thesis I proposed a game with the theme of evolution of the movement of virtual creatures, I created a system for their evolution based on information from existing works and expanded it with the features needed for the proposed game. In the course of the work, it turned out that the evolution is too computationally demanding to use in the game. Therefore, I created a simplified game prototype and in my work I focused more on experimenting with various settings of the algorithms for the evolution of virtual creatures.
|
guest ::
RSS feed.