|
|
Optimization of control using reinforcement learning on the Robocode platform
Pastušek, Václav ; Myška, Vojtěch (referee) ; Burget, Radim (advisor)
This master's thesis focuses on optimizing the control of a tank robot in the Robocode environment using reinforcement learning. The complexity of this problem falls into the EXPSPACE class, presenting a challenge that cannot be underestimated. The theoretical part of the thesis meticulously examines the Robocode platform, concepts of reinforcement learning, and relevant algorithms, while the practical part focuses on optimizing the agent, implementing reinforcement learning algorithms, and creating a user-friendly interface for easy training and testing of models. A total of 64 models were trained and tested as part of the thesis, with their data and parameters compared and presented in accompanying databases and graphs. The best results in terms of average hits per episode were achieved by models labeled v0.8.0 and v1.0.0. The first model exhibited a certain ability to evade shots, while the second model showed more successful hits.
|
|
|
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.
|
|
|
Reinforcement Learning for 3D Games
Beránek, Michal ; Herout, Adam (referee) ; Hradiš, Michal (advisor)
Thesis deals with neural network learning on simple tasks in 3D shooter Doom, mediated by research platform ViZDoom. The main goal is to create an agent, which is able to learn multiple tasks simultaneously. Reinforcement learning algorithm used to achieve this goal is called Rainbow, which combines several improvements of DQN algorithm. I proposed and experimented with two different architectures of neural network for learning multiple tasks. One of them was successful and after a relatively short period of learning it reached almost 50% of maximum possible reward. The key element of this achievement is an Embedding layer for parametric description of task environment. The main discovery is, that Rainbow is able to learn in 3D environment and with the help of Embedding layer, it is able to learn on multiple tasks simultaneously.
|
|
|
Navigation Using Deep Convolutional Networks
Skácel, Dalibor ; Veľas, Martin (referee) ; Hradiš, Michal (advisor)
In this thesis I deal with the problem of navigation and autonomous driving using convolutional neural networks. I focus on the main approaches utilizing sensory inputs described in literature and the theory of neural networks, imitation and reinforcement learning. I also discuss the tools and methods applicable to driving systems. I created two deep learning models for autonomous driving in simulated environment. These models use the Dataset Aggregation and Deep Deterministic Policy Gradient algorithms. I tested the created models in the TORCS car racing simulator and compared the result with available sources.
|
|
|
Utilization of Robotic Operating System (ROS) for control of collaborative robot UR3
Juříček, Martin ; Matoušek, Radomil (referee) ; Parák, Roman (advisor)
The aim of the bachelor's thesis is to create a control program, its subsequent testing and verification of functionality for the collaborative robot UR3 from the company Universal Robots. The control program is written in python and integrates control options through the Robotic Operating System, where a defined point can be reached using pre-simulated trajectories of Q-learning, SARSA, Deep Q-learning, Deep SARSA, or using only the MoveIT framework. The thesis deals with a cross-section of the topics of collaborative robotics, Robotic Operating System, Gazebo simulation environment, feedback and deep feedback learning. Finally, the design and implementation of the control program with partial parts is described.
|
|
|
Reinforcement learning for solving game algorithms
Daňhelová, Jana ; Uher, Václav (referee) ; Kolařík, Martin (advisor)
The bachelor thesis Reinforcement learning for solving game algorithms is divided into two distinct parts. The theoretical part describes and compares the fundamental methods of reinforcement learning with special attention to the methods of active learning – Q-learning and deep learning. In the practical part the deep q-learning technique is chosen for testing and applied to the case of the Snake game. The results are presented in the form of program written in Python programming language, which consists of the game environment created in PyGame, the model of convolutional neural network designed in Keras and agent playing the game. As an output of the program there are several types of datasets in CSV format. The gained data containing the values of parameters like number of epochs, accuracy, loss or the amount of the reward can later be used for further processing.
|
|
|
Navigation Using Deep Convolutional Networks
Skácel, Dalibor ; Veľas, Martin (referee) ; Hradiš, Michal (advisor)
This thesis studies navigation and autonomous driving using convolutional neural networks. It presents main approaches to this problem used in literature. It describes theory of neural networks and imitation and reinforcement learning. It also describes tools and methods suitable for a driving system. There are two simulation driving models created using learning algorithms DAGGER and DDPG. The models are then tested in car racing simulator TORCS.
|
|
|
Robocode - secured platform for evaluation of students' projects
Peňáz, Vladimír ; Ježek, Štěpán (referee) ; Burget, Radim (advisor)
This bachelor's thesis focuses on the design and implementation of a secure testing platform based on the game Robocode, which is used for evaluating student projects in the MSC-PDA subject. The project utilizes principles of machine learning and addresses a problem in the complexity class EXPSPACE. Evaluating the quality of results in this complexity class is challenging, and currently, there is no suitable environment available for these purposes. The objective of this thesis is to create a secure environment that allows students to compete on a game server with minimal risk of damaging the teacher's computer and ensures superuser privileges. Students will connect their trained models to the game server, where they will receive complete information about the battlefield, based on which they generate instructions for their tanks. In this way, the model will have the same information about the battle as a manually playing human. Based on the final score, it will be possible to evaluate which model performed the best. The platform is implemented in Java and works with models implemented in Python.
|
|
|
Using reinforcement learning to learn how to play text-based games
Zelinka, Mikuláš ; Kadlec, Rudolf (advisor)
The ability to learn optimal control policies in systems where action space is defined by sentences in natural language would allow many interesting real-world applications such as automatic optimisation of dialogue systems. Text-based games with multiple endings and rewards are a promising platform for this task, since their feedback allows us to employ reinforcement learning techniques to jointly learn text representations and control policies. We present a general text game playing agent, testing its generalisation and transfer learning performance and showing its ability to play multiple games at once. We also present pyfiction, an open-source library for universal access to different text games that could, together with our agent that implements its interface, serve as a baseline for future research.
|
|
|
Stock Trading Using a Deep Reinforcement Learning and Text Analysis
Benk, Dominik ; Baruník, Jozef (advisor) ; Vácha, Lukáš (referee)
The thesis focuses on exploiting imperfections on the stock market by utilizing state-of-the-art learning methods and applying them to algorithmic trading. The automated decisions are expected to have the capability of outperforming professional traders by considering much more information, reacting almost instantly and being unaffected by emotions. As an alternative to traditional supervised learning, the proposed model of reinforcement learning employs a principle of trial-and-error, which is essential for learning behaviours of all organisms. In the context of stocks, this allows to consider the involved uncer- tainty and therefore more precisely estimate the long-run returns. To collect the most relevant information for each trading decision, additionally to tech- nical indicators the models build on investor's opinion - financial sentiment. This is derived from two textual sources, news and social media, and the main goal is to compare their relative contribution to trading. Models are applied to 11 different stocks and later combined into portfolio for greater robustness of results. The textual analysis proves to be important for the learning process, especially in case of stocks with good media coverage. The Twitter is found to provide more valuable information compared to news, but their...
|
guest ::
