-
Franciszek Pelc (Warsaw University of Technology)18/04/2024, 09:30
Internet intrusion detection systems (IDS) use machine learning models, which one needs to train using public datasets. The training process requires a training set, which is a majority part of such a dataset, while validation is performed on its second part - the validation set. Finally, to evaluate the quality of the output model, one utilizes the test set, which is the third part. The...
Go to contribution page -
Michał Balas18/04/2024, 09:42
This paper proposes a solution to the problem of automatically generating an efficient front-end layer code based on a provided UI design. With the increasing complexity and scalability requirements of systems, companies, as well as individual developers, are inclined to seek a sufficient and maintainable way to automate some internal processes. One such automation may involve focusing on...
Go to contribution page -
Jakub Stolarski, Piotr Olechno, Mateusz Gietka18/04/2024, 09:54
The paper explores different solutions for implementing self-learning artificial intelligence (AI) competitive bots for the game Blood Bowl. The winners of the most of the previous competitions were scripted bots but in recent years bots based on machine learning started to outspace their competition. Blood Bowl is a two-player, turn-based, asymmetric board game that combines elements of...
Go to contribution page -
Aleksandra Kowalczyk18/04/2024, 10:06
The paper discusses the significance of one-shot learning from prototype SKU images for efficient product recognition in various retail and inventory management sectors. Traditional methods require large supervised datasets for training deep neural networks, which can be costly and impractical. One-shot learning techniques address this issue by enabling classification from a single prototype...
Go to contribution page -
Bartosz Sadowski (Warsaw University of Technology)18/04/2024, 10:18
Procedural plot generation is a topic widely researched in the context of video games. This paper discusses parts of the existing research using Role Playing Games as a target for plot generation. Analysis of table-, atom-, and Large Language Model-based approaches to plot generation for Role Playing Games indicates that more is needed. This paper proposes the solution to this problem using a...
Go to contribution page -
Ms Natalia Bernardelli
This paper investigates the process discovery conducted using process mining on anonymized logs extracted from the university's departmental application handling system (ISOD) in CSV file format. The study focuses on analyzing the process occurring within the application handling system. Activities and analyses are based on logs from this system. The proper data processing was needed to allow...
Go to contribution page -
Kamil Grzegorzewski
The presented work utilizes machine learning methods to solve the problem of detecting epileptic seizures using EEG signals. Epilepsy is one of the most common neurological diseases, affecting millions of people worldwide. This disease has always been of great importance in the field of biomedicine due to the health risks it poses. It is characterized by recurring, unprovoked epileptic...
Go to contribution page -
10. Automated Identification of Malignant Breast Lesions: A Convolutional Neural Network PerspectiveMateusz Kozicki
Accurate diagnosis and prognosis are crucial for the effective treatment of breast cancer. To improve diagnostic accuracy and reduce human error, this study presents a novel approach using deep neural networks (DNNs) to detect cancerous lesions in microscopic specimens from the breast. In this paper, I examine the performance of convolutional neural network (CNN) model as well as CNN-based...
Go to contribution page -
Justyna Budzyńska, Maria Kujawa
In recent years, deep neural networks (DNNs) have shown remarkable potential in medical image analysis, particularly in the field of computed tomography (CT) imaging. The use of DNNs to analyze medical images, especially in the context of detecting cancer, is becoming a promising area of research. In this study a novel approach to detect liver cancer using deep neural networks (DNNs) based on...
Go to contribution page -
Mr Adrian Mostowski, Bartłomiej Kopyść, Mr Kacper Kuczewski
Abstract:
In software development, clean code and clean architecture are crucial aspects that
Go to contribution page
ensure separation between business logic, application logic, and framework-related
code. However, in the dynamic world of web development, these approaches are not
commonly utilized due to the lack of standardization among frontend frameworks and
libraries. This often leads to complications in... -
Bartłomiej Anczok (Warsaw University Of Technology)
Advancements in UNet architectures have been pivotal in medical image segmentation, particularly for blood vessel segmentation, which is crucial for medical diagnosis and treatment. This article presents a comprehensive comparative study of various UNet models, examining their effectiveness in blood vessel identification within medical imaging. We explore the evolution of these models from the...
Go to contribution page -
Kajetan Jeznach
Deep neural networks (DNNs) are becoming a handy tool in the healthcare field. Research work in recent years has led to the creation of solutions that can effectively support the work of medical staff. This paper presents a comparison of deep learning architectures used for the classification of cardiovascular diseases. Models were trained and tested on the PTB-XL dataset, a large publicly...
Go to contribution page -
Aleksandra Koźbiel, Olgierd Usidus
This research presents a comparative analysis of
Go to contribution page
Natural Language Understanding (NLU) and Natural Language
Generation (NLG) models for the task of fake news detection. A
concise literature review was conducted to understand the state-
of-the-art techniques in the field. The study focused on comparing
the performance of two language models, BERT (Bidirectional
Encoder Representations from... -
Adrian Krzemiński, Hubert Kunikowski, Szymon Maliszewski
This scientific article describes a project focused on creating a convolutional neural network designed to assist physicians in diagnosing diabetic retinopathy based on Optical Coherence Tomography (OCT) images. The project involves an in-depth analysis of existing research on the classification of this disease using fundus images. Leveraging a diverse dataset of OCT images, encompassing both...
Go to contribution page -
Bartosz Obstawski, Wojciech Nowicki
As artificial intelligence (AI) systems increasingly penetrate various domains, the need to understand their decision-making processes becomes paramount. This study presents a comprehensive review of Python-based eXplainable AI (XAI) packages aimed at deciphering the black box nature of AI models. By analyzing and comparing prominent XAI techniques and tools, this review sheds light on their...
Go to contribution page -
Kamil Kowieski, Kinga Kocoł
In the era of rapidly advancing technology, the spread of deepfake content has become an escalating challenge, making detecting synthetically created materials crucial in softening the potential harms. This paper focuses on a comparative analysis of Convolutional Neural Networks (CNNs) designed explicitly for deepfake detection (MesoNet and MesoNet Inception) with a custom architecture...
Go to contribution page -
Agnieszka Salwa, Dawid Szczawiński, Hubert Skibiński
Contemporary medical diagnostics increasingly utilise advanced information technologies to analyse microscopic specimens, crucial in detecting and diagnosing cancerous changes. In our work, we focused on the analysis of microscopic specimens from the colon, which is extremely important in detecting early stages of cancer. We used Python to process and analyse these images, offering extensive...
Go to contribution page -
KIatarzyna Prus
This paper embarks on a comparative study of three diverse reinforcement learning techniques applied to forecasting and optimizing energy trading in day-ahead markets for medium-sized prosumers. Given renewable energy sources' inherent volatility and unpredictability, this study leverages these diverse approaches, each known for its unique advantages in navigating complex optimisation...
Go to contribution page -
Mr Eryk Sędkowski, Wojciech Zieliński
This paper aims to improve existing solutions of machine-learned bots for the Blood Bowl game. Blood Bowl is a stochastic, fully-observable, turn-based, two-player board game. Players, referred to as coaches, lead an 11-player team, where each individual has statistics and abilities. A wide grid-based board, various moves, possible interruption of the game sequence, and randomness create a...
Go to contribution page -
Karolina Gałczyńska
This paper uses a prototyping network to address the problem of recognizing products on store shelves. The problem of identifying store products is similar to the problem of facial recognition. The number of facings is practically unlimited, and new products are introduced every week, current labels are changed, and new product variants are added. Collecting appropriate training patterns is...
Go to contribution page -
Maciej Dragun
Sense of sight is fundamental in one's life and one can hardly imagine living without it yet in modern times eye diseases can develop relatively early and without timely reaction can lead to severe eyesight degradation or even loss. A means widely used in ophthalmology is fundus photography which can be obtained with relativaly low effort. Their analysis can reveal symptoms of various...
Go to contribution page -
Dominik Biedrzycki (Politechnika Warszawska)
In this comprehensive exploration, the author delves
Go to contribution page
into a visionary initiative aimed at fundamentally transforming
language learning through the implementation of a groundbreaking Conversational AI system. The research underscores the
pivotal role of innovative features, particularly the successful
integration and testing of TextToSpeech and real-time translations, which have emerged as... -
Ms Agata Lachowiecka, Mr Maxymilian Kowalski, Patryk Guba
Finite-Difference Time-Domain method of electromagnetic field computation often requires significant amount of time and memory resources, which emphasizes the necessity for the development of high-performance programs. In this paper, we present enhancements to the performance of an electromagnetic field solver using the FDTD method on an asymmetric grid. The improvement in efficiency was...
Go to contribution page -
Mohamad Al-Tabich
Since the development of advanced gaming engines, pathfinding has evolved into a crucial role in enhancing the complexity and realism of non-player characters (NPCs). This article explores the implementation of the D* algorithm in Unity, a leading solution in game development. It features a comparison of various notable pathfinding algorithms, tested in a generated labyrinth with interactable...
Go to contribution page -
Mr Piotr Szczepański, Mr Karol Zieliński, Mr Albert Ziółkiewicz
The research focused on classic image captioning based on a coder-decoder structure, where the coder encodes the image features. At the same time, the decoder produces a caption – a phrase describing the image content. We investigated the decoder part by testing multiple convolutional-neural-network-based backbones – feature extractors. This investigation aimed to find the optimal encoder,...
Go to contribution page -
Stanisław Świder
Precise prediction of photovoltaic (PV) energy generation is essential for optimal, profitable and ecological management of electric energy resources all over the world. As a result, one attempts to develop more accurate prediction algorithms. This paper compares the application of Long Short-Term Memory (LSTM), a subtype of Recurrent Neural Networks, with Transformer Neural Network for...
Go to contribution page -
Agata Biernacka, Mr Kacper Kilianek
Pulmonary hypertension (PH) refers to a group of diseases characterized by elevated blood pressure in the pulmonary arteries supplying the lungs. Accurate diagnosis of PH and determining its underlying causes is crucial but often challenging for physicians. The development of artificial intelligence (AI) techniques may help address this challenge by enhancing the diagnostic process. This paper...
Go to contribution page -
Barbara Gałczyńska, Michał Wyrostkiewicz (Warsaw University of Technology)
Quantum computers promise to revolutionize several fields, including cryptography. In recent years, researchers have made significant progress in developing quantum algorithms that can solve computational problems much faster than classical computers. These advances have led to concerns about the security of traditional cryptography algorithms, as they may be vulnerable to quantum attacks. In...
Go to contribution page -
Maciej Konieczny
Maxwell equations are fundamental laws of physics for understanding electromagnetic fields. Their usage is essential in applications related to propagation of electromagnetic waves. To provide a numerical implementation of these equations, a Finite-Difference Time-Domain method can be used. It is crucial to optimize such complex calculations in order to provide convenient, time-saving tools....
Go to contribution page -
Maciej Bukalski
The paper presents research on the usefulness of the A algorithm in the context of setting a route for an overhead power line. The algorithm was implemented in Python using the pyautocad library. The final result of the program is the marked route in a dwg file. The research examined how different ways of implementing the A algorithm and different heuristics affects its performance. The...
Go to contribution page -
Damian Janczarek
This article conducts a comprehensive comparison of Mistral7B and SQLCoder 2.0, two „small” large language models, in the context of the Text2SQL task using the „Spider” dataset. Despite its modest scale, Mistral7B achieves a notable 33% accuracy without any query-answer examples in the prompt, showcasing promising prospects for compact large language models in the Text2SQL domain. The...
Go to contribution page -
Antoni Malinowski, Arkadiusz Kasprzak
This study introduces a microservice-based architecture designed to compute a large amount of data, offering scalable and modular system components. Main system concepts are illustrated through the example of credit risk management in a fictional banking system. The architecture's design principles emphasize scalability, domain separation, and technological agnosticism, aiming to ensure...
Go to contribution page -
Mr Jakub Matłacz (Warsaw University Of Technology, Faculty Of Electrical Engineering)
This paper presents a novel approach, Analyze-Select-Match (ASM), for local email categorization using small language models. The objective is to enable users to organize emails on their local machines by categorizing them into user-defined labels with flexibility in both quantity and quality. A dataset of email samples was curated, and five models, sized for widespread graphics card...
Go to contribution page -
Mr Bartosz Sieracki
Simulation environments perform an important function in modern technical processes. Skillful utilization of such environments reduces testing costs and time associated with implementing solutions, thereby expediting development efforts. This paper focuses on reviewing the application of selected simulation environments for integration with the JetRacer ROS AI KIT mobile platform. This study...
Go to contribution page
Choose timezone
Your profile timezone: