|
|
Reprezentace síťových toků s využitím neuronových sítí
Pycz, Lukasz ; Jeřábek, Kamil (referee) ; Poliakov, Daniel (advisor)
This thesis explores the application of self-supervised learning (SSL) methods such as data masking, data order shuffling, and contrastive learning, to extract meaningful representations from network flow data, specifically using the CESNET TLS22 dataset from CESNET DataZoo. The main goal is to develop a robust model that improves the understanding and analysis of network flows through effective representation learning without relying on labeled data. The research utilizes the PyTorch computational framework for designing, training, and evaluating the performance of the model.
|
|
|
Simulation of Biological Processes Using Asynchronous Cellular Automata and Machine Learning
Kališ, Vojtěch ; Bidlo, Michal (referee) ; Fritz, Karel (advisor)
Tato práce zkoumá spojení asynchronních celulárních automatů a technik strojového učení pro simulaci komplexních biologických procesů. Jejím hlavním zaměřením je předvést vrozený potenciál výpočetního rámce konstruovaného spojením paralelismu aktualizačního modelu asynchronních celulárních automatů s prediktivními schopnostmi algoritmů strojového učení. Tato studie si klade za cíl demonstrovat kvality takového hybridního přístupu implementací tří matematických modelů celulárních automatů s rostoucí složitostí—tj., seřezeny podle stupně složitosti, Conwayova Hra Života, SmoothLife a Lenia—ve své základní formě a následnou integrací strojového učení do funkce dvou posledně jmenovaných, po čemž následuje porovnání výsledků obou přístupů.
|
|
|
Stereo Reconstruction with Deep Neural Networks
Letanec, Richard ; Herout, Adam (referee) ; Španěl, Michal (advisor)
The aim of this thesis is to design and train a neural network model capable of estimating a disparity map from a pair of images. It will then be possible to create a depth map and point cloud from the estimated disparity map. Such a process is called stereo reconstruction. Solving this task consists of two steps -- choosing a suitable dataset and choosing a suitable neural network architecture. In my work, I compared two neural network architectures that I trained on the DrivingStereo dataset, consisting of paired images photographed from the roof of a car, and retrained and evaluated on the KITTI 2015 dataset, consisting of images of the same type. As the first neural network architecture, I chose ES-Net, which uses an approach based on a sequence of residual blocks and convolutional layers. As the second architecture, I chose CREStereo, which uses an iterative approach based on recurrent layers to predict the disparity map. In all benchmark tests, the CREStereo architecture achieves better accuracy.
|
|
|
Data augmentation integration into Pytorch
Vašina, Ladislav ; Polok, Alexander (referee) ; Szőke, Igor (advisor)
This thesis presents a tool that creates a unified, simple, and user-friendly interface on top of the audio augmentation libraries that can be used in conjunction with PyTorch library. The implemented tool offers the possibility to use a wide spectrum of augmentations from different libraries and offers easy application of those augmentations on the datasets. The support of the large range of augmentations could be only achieved by using multiple interfaces of the individual libraries. The tool can receive a list of augmentations from the user with its parameters and then it decides which of the integrated libraries it should use to apply that specific augmentation. The created tool was tested on the task of fine-tuning the automatic speech recognition system called Whisper. The main contribution of this work is that it provides a solution to a large number of libraries for the augmentation of audio data, where each library provides a different number and types of augmentations of audio, while also having different features and interfaces.
|
|
|
Tracking people based on their clothing in multi-camera systems
Sivak, Mykyta ; Přinosil, Jiří (referee) ; Číka, Petr (advisor)
This bachelor thesis focuses on the development and implementation of an algorithm for tracking individuals in multi-camera systems based on clothing pattern analysis. The aim was to design a system capable of tracking an individual in various positions and frames, using the Region of Interest (RoI) technique. The study begins with a comprehensive review of the existing literature on object tracking in video sequences, with a special focus on RoI tracking techniques. During the research, a new algorithm was developed and implemented that utilizes clothing patterns as the primary identification element for tracking and re-identifying individuals across different camera shots. The algorithm was experimentally validated on datasets containing video sequences from various environments, allowing for a detailed analysis of its effectiveness and reliability. The experimental results demonstrate that the proposed system achieves significant accuracy and efficiency compared to traditional methods and is particularly effective in challenging situations where other methods fail. The thesis concludes with an evaluation of the conducted experiments along with recommendations for future extensions and improvements of the system. Potential challenges and ethical aspects, including issues of privacy and personal data processing, are also discussed.
|
|
|
Neural Networks at the Level of Network Packets and Flows
Urbánek, Petr ; Jeřábek, Kamil (referee) ; Poliakov, Daniel (advisor)
Tato diplomová práce se zabývá integrací neuronových sítí do monitorování toků v síti, zejména se zaměřením na ipfixprobe — open-source exportér IP toků sítí vyvinutý společností CESNET. Cílem je zkoumat potenciál neuronových sítí pro klasifikaci a extrakci reprezentací ze síťových toků. Jsou zde zvažovány výzvy spojené s nasazením takových řešení ve velkém měřítku v produkčních prostředích, s konkrétním důrazem na zlepšení efektivity a účinnosti v dynamickém technologickém prostředí.
|
|
|
Identification of specified segments in the audio signal using machine learning
Pařízek, Radim ; Galáž, Zoltán (referee) ; Zvončák, Vojtěch (advisor)
The bachelor thesis deals with the design of a system for the identification of natural environmental sounds in audio recordings. The datasets and models used for this type of tasks are surveyed and their structure is described. A system for the identification of sounds in one layer and in two layers has been proposed for seven selected labels. The classifier used for this system was created by fine-tuning a transformer model from the Hugging Face platform. The results of two training approaches and one identification system were evaluated.
|
|
|
Device for the traffic situation evaluating
Gábel, Matej ; Honec, Peter (referee) ; Janáková, Ilona (advisor)
The bachelor's thesis deals with the implementation of a device for evaluating the traffic situation, specifically by traffic signs detection. In this work, I tried different methods of traffic sign recognition, where the resulting implementation on hardware is done using convolutional neural networks. More precisely, it is the YOLOv5 architecture, which is suitable for recognizing traffic signs in real time.
|
|
|
Self-supervised learning in computer vision applications
Vančo, Timotej ; Richter, Miloslav (referee) ; Janáková, Ilona (advisor)
The aim of the diploma thesis is to make research of the self-supervised learning in computer vision applications, then to choose a suitable test task with an extensive data set, apply self-supervised methods and evaluate. The theoretical part of the work is focused on the description of methods in computer vision, a detailed description of neural and convolution networks and an extensive explanation and division of self-supervised methods. Conclusion of the theoretical part is devoted to practical applications of the Self-supervised methods in practice. The practical part of the diploma thesis deals with the description of the creation of code for working with datasets and the application of the SSL methods Rotation, SimCLR, MoCo and BYOL in the role of classification and semantic segmentation. Each application of the method is explained in detail and evaluated for various parameters on the large STL10 dataset. Subsequently, the success of the methods is evaluated for different datasets and the limiting conditions in the classification task are named. The practical part concludes with the application of SSL methods for pre-training the encoder in the application of semantic segmentation with the Cityscapes dataset.
|
|
|
Neural Networks for Network Anomaly Detection
Matisko, Maroš ; Martinásek, Zdeněk (referee) ; Blažek, Petr (advisor)
This bachelor thesis is focused on creating a system to mitigate computer network attacks. One of the most common groups of attacks is Distributed Denial of Service (DDoS) attacks, against which this system should protect internal network. In the theoretical part of the thesis are described DDoS attacks, existing systems for their mitigations, neural networks principle and their use. Practical part consists of choosing communication parameters, constructing a neural network with use of these parameters, implementation of this neural network in real–time attack mitigation system and a result of testing of this system.
|
guest ::
