Student IT/EE Workshop 2025

Session

Session B (Poster)

24 Apr 2025, 10:45

SK 108 (Stara Kotłownia)

31. A Lung Volume-Based Perfusion SPECT and CT Comparison Algorithm for Enhanced Pulmonary Embolism Diagnosis

Mr Daniel Kisiel (Politechnika Warszawska), Mr Eryk Muchorowski (Politechnika Warszawska)

24/04/2025, 10:45

Poster

I. BACKGROUND
Pulmonary Embolism (PE) is reported to be one of the most
common cardiovascular diseases. It is caused by a blood clot
that develops in a blood vessel elsewhere in the body and
travels to an artery in the lung, forming a blockage [1].

Lung V/Q (ventilation/perfusion) SPECT is one of an es-
tablished diagnostic imaging test for suspected PE. The idea
behind this test is...

2. Accelerating AI Inference in the Browser with WebGPU: Evaluating Quantization Trade-offs in Latency, Quality, and Memory Usage

Mr Ignacy Ruszpel (Politechnika Warszawska - Wydział Elektryczny), Mr Nikodem Wójcik (Politechnika Warszawska - Wydział Elektryczny)

24/04/2025, 10:45

Poster

Recent advances in deep learning and natural language processing have spurred the demand for deploying increasingly complex models on resource-constrained platforms. Modern browser environments, empowered by emerging GPU standards like WebGPU, now offer a promising venue for real-time AI inference. This paper provides an overview of leveraging WebGPU for accelerating inference directly within...

21. Analysis and comparison of missing value imputation methods for atmospheric pollution data

Jakub Jasiński (Warsaw University of Technology)

24/04/2025, 10:45

Poster

Missing values are a common phenomenon in real-world time series datasets and can significantly impact the precision and reliability of data analysis and machine learning models. This research project aims to discuss the types of missing data occurrence and test and analyze different possibilities of their imputation. The methods taken into consideration will start from the simplest ones based...

41. Automatic translation of Japanese manga using MultiModal Large Language Models

Tomasz Nitsch (Warsaw University of Technology)

24/04/2025, 10:45

Poster

Japanese manga has captivated readers worldwide with its vibrant and expressive form of art that provides compelling storytelling with intricate visuals. However, for many fans outside of Japan, language barriers often stand in the way of fully experiencing the depth of these stories. Traditional translation from language to language, while effective, can be a very time-consuming and...

23. Comparative Analysis of Machine Learning and Statistical Models for Short-Term Energy Production Forecasting in Poland

Jakub Karpiński (Warsaw University of Technology)

24/04/2025, 10:45

Poster

This paper investigates short-term energy produc-
tion forecasting in Poland using three distinct predictive mod-
els: LightGBM, an ARIMA-based model, and a Long Short-
Term Memory (LSTM) network. Leveraging historical data
from Poland’s energy sector, our study evaluates each model’s
performance in terms of accuracy, robustness, and computational
efficiency. The LightGBM model employs...

17. Comparative Analysis of Multi-Agent LLM Systems for Solving Polish Matura in Physics Exams

Piotr Wróbel (Politechnika Warszawska)

24/04/2025, 10:45

Poster

Large Language Models have gained widespread
recognition since OpenAI released their revolutionary model,
ChatGPT 3.5. Since then, many new approaches have emerged
to improve the capabilities and accuracy of these models for
different tasks. One such method involves using multi-agent
conversations. This article compares two multi-agent setups
designed to solve the Polish standardized...

30. Comparing the Efficiency of Selected Reinforcement Learning Algorithms in Stability Control and Navigation Tasks

Oskar Wyłucki

24/04/2025, 10:45

Poster

This paper presents a comprehensive comparison of the ef-
ficiency of four key reinforcement learning algorithms (DQN,
PPO, REINFORCE, and A2C) in stability control and nav-
igation tasks. The study was conducted in two test environ-
ments: Cart Pole, representing a basic balance maintenance
task, and Lunar Lander, constituting a complex navigational
challenge requiring precise landing....

34. Comparison of Models for Automatic Description of Medical Images

Mr Jakub Urbański (Warsaw University of Technology)

24/04/2025, 10:45

Poster

Automated medical report generation from chest X-ray images is a critical area of research in medical AI, aiming to enhance diagnostic accuracy, reduce radiologists' workload, and improve patient care. The process involves analyzing medical images and translating visual findings into structured, clinically relevant textual reports. Traditional methods rely on human expertise, which is...

48. Enhancing Large Language Models with Retrieval-Augmented Generation: A Case Study on Movie Data Beyond the Training Cutoff

Marcel Mikołajczyk (Politechnika Warszawska, Wydział Elektryczny)

24/04/2025, 10:45

Poster

This article investigates the role of Retrieval-Augmented Generation (RAG) in enhancing Large Language Models (LLMs) with information about movies and TV series released beyond their training data. In this study, the Llama 3.2 3B LLM is leveraged and integrated with external movie-related data retrieved from the OMDb API to provide specific information about over 14000 titles released in 2024,...

10. Enhancing Phishing Detection on Websites: A Hybrid Approach Combining Signature-Based Detection and Traditional Machine Learning Methods - Abstract

Jakub Romanek (Warsaw University of Technology), Marcin Latawiec

24/04/2025, 10:45

Poster

With the increasing popularity of web services and the significant amount of time users spend online, cybercriminals are increasingly targeting this space with sophisticated malicious techniques. Phishing, one of the most prevalent cybersecurity threats, poses a substantial risk to both individuals and organizations. These attacks are typically executed via deceptive emails containing...

50. Ethereum price forecasting using deep neural networks

Oskar Rozwadowski (Warsaw University of Technology)

24/04/2025, 10:45

Poster

Short-term forecasting of cryptocurrency prices remains a challenging task due to the high volatility and complex market dynamics of digital assets like Ethereum. This study proposes a hybrid deep learning model, that integrates networks such as Bi-LSTM, FinBERT and GRU, seeking to provide a comprehensive analysis of their applicability in this domain and enhance predictive accuracy for this...

16. Evaluating detection methods of synthetic images

Mr Michal Wójtowicz (Warsaw University of Technology), Michał Biliński (Warsaw University of Technology)

24/04/2025, 10:45

Poster

The paper aims to evaluate the effectiveness of available detectors in distinguishing fake from real photos. Deepfakes are artificially generated images or videos created using artificial intelligence or manual tools like Photoshop. New techniques, such as Generative Adversarial Networks (GANs) and Diffusion Models (DMs) enable the rapid generation of highly realistic images. The research...

29. Evaluating unsupervised data mining methods to assess the utility of such approaches for SIEM event analysis.

Marcin Dydo (Warsaw University of Technology)

24/04/2025, 10:45

Poster

Modern organizations generate vast amounts of data, a significant portion of which consists of system, network, and application logs. The sheer volume and scalability of these logs make manual analysis inefficient and highly time-consuming. Automated anomaly detection techniques have been developed and refined to accelerate this process. Currently it is common for open SIEM systems to only...

19. Examination of PCA Utilisation for Multilabel Classifier of Multispectral Images

Mr Bartosz Staszynski (Warsaw University of Technology), Mr Filip Karpowicz (Warsaw University of Technology), Wiktor Kępiński (Warsaw University of Technology)

24/04/2025, 10:45

Poster

This paper investigates the utility of Principal Component Analysis (PCA) for multi-label classification of multispectral images using ResNet50 and DINOv2, acknowledging the high dimensionality of such data and the associated processing challenges. Multi-label classification, where each image may belong to multiple classes, adds further complexity to feature extraction. Our pipeline includes...

15. Explainability of Convolutional Neural Network: Overview of Methods

Mr Magdalena Markowicz (PW)

24/04/2025, 10:45

Poster

Understanding and interpreting the decisions made by deep learning models has become an essential area of research in artificial intelligence. Convolutional neural networks (CNNs), despite their high performance in various tasks, often function as "black boxes," making it challenging to explain their predictions. This study focuses on applying and evaluating different explainability techniques...

39. Exploring LLMs mathematical reasoning capability: Insights from GSM-Symbolic in English and Polish

Marcin Łomiński (Wydział Elektryczny, Politechnika Warszawska), Michał Tomczyk (Wydział Elektryczny, Politechnika Warszawska)

24/04/2025, 10:45

Poster

Large language models (LLMs) are trained with ever-increasing amounts of data. It seems that when asked to solve mathematical tasks, they can infer and reason mathematically [1]. The GSM8K benchmark platform is widely used to test various LLMs to solve simple arithmetic tasks. Recently, LLMs have shown a clear improvement in their ability to correctly answer questions from the GSM8K dataset....

40. Franky: An Intelligent Agent for Stock Portfolio Management Using Large Language Models and Deep Reinforcement Learning

Mateusz Bartosik (Warsaw University of Technology), Mikolaj Zawada (Warsaw University of Technology)

24/04/2025, 10:45

Poster

Franky: An Intelligent Agent for Stock Portfolio
Management Using Large Language Models and
Deep Reinforcement Learning
1st Mikołaj Zawada∗
Faculty of Electrical Engineering
Warsaw University of Technology
Warsaw, Poland
mikolaj.zawada.stud@pw.edu.pl
2nd Mateusz Bartosik∗
Faculty of Electrical Engineering
Warsaw University of Technology
Warsaw,...

45. Optimizing Football Betting Strategies Using Deep Neural Networks and Modern Portfolio Theory

Daniel Kuciński, Mr Tomasz Mycielski (Warsaw University of Technology)

24/04/2025, 10:45

Poster

This paper presents a hybrid framework combining Deep Neural Networks (DNNs) and Modern Portfolio Theory (MPT) to optimize football betting strategies. Leveraging historical match data from five major European leagues (2014–2025), we engineer predictive features such as dynamic Elo rankings, expected goals (xG), and team performance metrics derived from raw game statistics. Diverse neural...

51. Pose-Based Motion Analysis for Physical Exercise Quality Assessment

Wojciech Jakieła (Politechnika Warszawska)

24/04/2025, 10:45

Poster

This paper presents a comprehensive approach to the evaluation of physical exercise using skeletal estimation and motion analysis. The research begins with the acquisition of relevant datasets and a review of existing monocular pose estimation solutions. A key component of this work is modeling important features, including skeletal joint weighting and key angles between them, according to...

20. Post-Quantum Cryptography: Benchmarking ML-KEM Against RSA

Nina Łabęcka (Warsaw University of Technology)

24/04/2025, 10:45

Poster

The rise of quantum computing presents a significant threat to classical cryptographic systems, particularly those relying on hard mathematical problems such as integer factorization and discrete logarithms. This paper explores the impact of quantum computing on traditional cryptographic algorithms and the necessity of transitioning to quantum-resistant cryptography. We provide an overview of...

42. Proxy Methods in Image Recognition

Barbara Borkowska, Marcin Maciąg

24/04/2025, 10:45

Poster

This paper presents a comprehensive comparison of several state-of-the-art proxy loss methods for retail product recognition, focusing on accuracy, computational efficiency, embedding quality, and convergence behaviour. An extensive evaluation is performed on the Stanford Online Products benchmark dataset, which contains over 120,000 images covering 22,634 distinct product categories. The...

14. Przewidywanie wstrząsów wtórnych przy pomocy Sieci Neuronowych.

Mikołaj Klębowski

24/04/2025, 10:45

Poster

Silne trzesienia ziemi niosa ze soba ryzyko wystapi-
enia wstrzas´ow wt´ornych, kt´ore moga powodowa´c wieksze
zniszczenia ni˙z gł´owne trzesienie. Ze wzgledu na zło˙zono´s´c
tego zjawiska oraz gwałtowno´s´c jego przebiegu, prognozowanie
czasu wystapienia i intensywno´sci wstrzas´ow wt´ornych stanowi
wcia˙z istotne wyzwanie dla sejsmologii. W niniejszym artykule
zaprezentowano podej´scie...

47. Testy penetracyjne systemu zarządzania energią elektryczną OpenEMS — identyfikacja i mitygacja podatności w otwartoźródłowych systemach energetycznych

Kacper Gorlowski

24/04/2025, 10:45

Poster

W odpowiedzi na rosnącą popularność rozproszonych systemów zarządzania energią oraz integrację odnawialnych źródeł energii (OZE), w projekcie przeprowadzono kompleksową analizę bezpieczeństwa otwartoźródłowego systemu OpenEMS. System ten, napisany w języku Java, wspiera monitorowanie, kontrolę i optymalizację przepływów energii elektrycznej w instalacjach przemysłowych i prosumenckich. Celem...

26. Verification of the authenticity of bullion coins with a mobile application based on frequency analysis using an artificial neural network

Tomasz Gryczka

24/04/2025, 10:45

Poster

Counterfeiting of bullion coins is a serious issue for collectors, dealers and investors, highlighting the need for reliable and accessible authentication tools. This research presents a mobile application designed to authenticate bullion coins using audio analysis, offering an innovative and convenient solution. The app uses autoencoders, a form of artificial neural network, trained on...

35. Zastosowanie modeli GAN do usuwania artefaktów z sygnału EEG

Małgorzata Sokół (Wydział Elektryczny, Politechnika Warszawska)

24/04/2025, 10:45

Poster

Sygnał elektroencefalograficzny (EEG) może ulegać zakłóceniom spowodowanym przez aktywność elektryczną niezwiązaną bezpośrednio z pracą mózgu badanego. Zjawisko to określane jest jako problem artefaktów EEG, które dzielimy na techniczne i fizjologiczne. Do artefaktów fizjologicznych należą charakterystyczne fragmenty sygnału rejestrowane w wyniku ruchów gałek ocznych, mrugania, czy aktywności...