Differences
This shows you the differences between two versions of the page.
| Both sides previous revisionPrevious revisionNext revision | Previous revision | ||
| aira:start [2025/12/17 21:10] – [Schedule Autumn 2025] mtm | aira:start [2026/04/16 14:27] (current) – [Schedule Spring 2026] mtm | ||
|---|---|---|---|
| Line 19: | Line 19: | ||
| Contact for enrollment of the JU PhD students [[https:// | Contact for enrollment of the JU PhD students [[https:// | ||
| + | |||
| + | ===== Schedule Spring 2026 ===== | ||
| + | |||
| + | * **[RESEARCH TRACK] 2026.04.16**: | ||
| + | * Meeting link: | ||
| + | * Recording: [[https:// | ||
| + | * Presentation slides: {{: | ||
| + | |||
| + | * **[PHD TRACK] 2026.04.09**: | ||
| + | * Meeting link: [[https:// | ||
| + | * Recording: | ||
| + | * Presentation slides: | ||
| + | |||
| + | * **[PHD TRACK] 2026.03.26**: | ||
| + | * Meeting link: | ||
| + | * Recording: | ||
| + | * Presentation slides: | ||
| + | |||
| + | * **[PHD TRACK] 2026.03.19**: | ||
| + | * Meeting link: | ||
| + | * Recording: | ||
| + | * Presentation slides: | ||
| + | |||
| + | * **[RESEARCH TRACK] 2026.03.12**: | ||
| + | * Meeting link: | ||
| + | * Recording: | ||
| + | * Presentation slides: | ||
| + | |||
| + | * **[RESEARCH TRACK] 2026.03.05**: | ||
| + | * Meeting link: | ||
| + | * Recording: | ||
| + | * Presentation slides: | ||
| ===== Schedule Autumn 2025 ===== | ===== Schedule Autumn 2025 ===== | ||
| + | |||
| + | * **[PHD TRACK] 2026.01.29**: | ||
| + | * Meeting link: | ||
| + | * Recording: | ||
| + | * Presentation slides: | ||
| + | |||
| + | * **[PHD TRACK] 2026.01.22**: | ||
| + | * Meeting link: | ||
| + | * Recording: [[https:// | ||
| + | * Presentation slides: | ||
| + | |||
| + | * **[RESEARCH TRACK] 2026.01.15**: | ||
| + | * Meeting link: | ||
| + | * Recording: | ||
| + | * Presentation slides: | ||
| + | |||
| + | * **[PHD TRACK] 2026.01.08**: | ||
| + | * Meeting link: | ||
| + | * Recording: | ||
| + | * Presentation slides: | ||
| * **[RESEARCH TRACK] 2025.12.18**: | * **[RESEARCH TRACK] 2025.12.18**: | ||
| * Meeting link: | * Meeting link: | ||
| - | * Recording: | + | * Recording: |
| - | * Presentation slides: | + | * Presentation slides: |
| * **[PHD TRACK] 2025.12.11**: | * **[PHD TRACK] 2025.12.11**: | ||
| Line 35: | Line 87: | ||
| * Meeting link: [[https:// | * Meeting link: [[https:// | ||
| * Recording: | * Recording: | ||
| - | * Presentation slides: | + | * Presentation slides: |
| * **[RESEARCH TRACK] 2025.11.27**: | * **[RESEARCH TRACK] 2025.11.27**: | ||
| Line 501: | Line 553: | ||
| + | |||
| + | |||
| + | |||
| + | ==== 2026-04-16 ==== | ||
| + | <WRAP column 15%> | ||
| + | {{ : | ||
| + | </ | ||
| + | |||
| + | <WRAP column 75%> | ||
| + | |||
| + | **Speaker**: | ||
| + | |||
| + | **Title**: Explaining Models or Modelling Explanations? | ||
| + | |||
| + | **Abstract**: | ||
| + | Counterfactual explanations (CE) and algorithmic recourse (AR) have emerged as promising approaches towards explaining opaque machine learning models and empowering individuals affected by them. This seminar will explore unexpected challenges and new opportunities in this context and demonstrate how counterfactuals can be used to improve the trustworthiness of models . It will summarize some of the main findings of Patrick' | ||
| + | |||
| + | **Biogram**: | ||
| + | Patrick is a trained economist, computer scientist, and researcher. In his research, he has challenged long-standing paradigms in explainable AI, developed novel methods to make AI more trustworthy, | ||
| + | </ | ||
| + | <WRAP clear></ | ||
| + | |||
| + | ==== 2026-04-09 ==== | ||
| + | <WRAP column 15%> | ||
| + | {{ : | ||
| + | </ | ||
| + | |||
| + | <WRAP column 75%> | ||
| + | |||
| + | **Speaker**: | ||
| + | |||
| + | **Title**: A Time-Aware GitHub Mining Framework for Empirical Software Quality Studies. | ||
| + | |||
| + | **Abstract**: | ||
| + | This research focuses on designing an automated system to assess the quality of GitHub repositories based on a predefined quality model. The proposed approach evaluates repositories using a range of metrics, including commit history and its associated metadata (such as size, timestamps, and descriptions), | ||
| + | In addition, the system may incorporate further quality indicators, potentially drawing on established frameworks such as ISO/IEC 25010:2011, to provide a more comprehensive and standardized evaluation. | ||
| + | |||
| + | **Biogram**: | ||
| + | Software engineer with a background in the telecom domain, specializing in network management systems and Software Defined Networking (SDN). Experienced in performance engineering, | ||
| + | </ | ||
| + | <WRAP clear></ | ||
| + | |||
| + | ==== 2026-03-26 ==== | ||
| + | <WRAP column 15%> | ||
| + | {{ : | ||
| + | </ | ||
| + | |||
| + | <WRAP column 75%> | ||
| + | |||
| + | **Speaker**: | ||
| + | |||
| + | **Title**: Formal Grammar Transducers in medical image analysis and segmentation correction | ||
| + | |||
| + | **Abstract**: | ||
| + | Syntactic Pattern Recognition is data analysis approach stemming from formal grammars, formal languages and syntax analyser development. It’s particularly effective in analysing structures, both those found in natural world and in human-made artifacts. To this day many studies used SPR methods in diagnosis of medical subjects, like hearing impairments in neonates or in commercial field, like for electricity consumption forecast. | ||
| + | Current PHD candidate’s research focuses on medical image analysis for patients with oligodendroglioma brain cancer. Goal of the endeavour is to support medics job in detecting and contouring cancer changes in brain. The glioma images acquired by MRI means are 2d scans. The segmentation would therefore benefit from a method that would correct it to represent a coherent 3d structure. | ||
| + | |||
| + | **Biogram**: | ||
| + | After working for a short time supporting eCRF (electronic Case Report Form) for various medical trials, Mateusz continues education as a first year PhD student at Technical Computer Science, Jagiellonian University. He holds both Bachelor' | ||
| + | He professionally worked on 3D medical image presentation, | ||
| + | </ | ||
| + | <WRAP clear></ | ||
| + | |||
| + | ==== 2026-03-19 ==== | ||
| + | <WRAP column 15%> | ||
| + | {{ : | ||
| + | </ | ||
| + | |||
| + | <WRAP column 75%> | ||
| + | |||
| + | **Speaker**: | ||
| + | |||
| + | **Title**: Beyond Heatmaps: Explaining Time Series with Post-hoc Attribution Rules and Counterfactuals | ||
| + | |||
| + | **Abstract**: | ||
| + | While complex machine learning models excel in time series classification, | ||
| + | |||
| + | **Biogram**: | ||
| + | Maciej Mozolewski is a PhD Researcher at Jagiellonian University and a member of the GEIST research group led by Prof. Grzegorz J. Nalepa. His work centers on human-centered explainable AI (XAI) for dynamic data, including multivariate time series and explanation visualization. He focuses on post-hoc methods that bridge the gap between complex machine learning models and human-intelligible explanations, | ||
| + | </ | ||
| + | <WRAP clear></ | ||
| + | |||
| + | ==== 2026-03-12 ==== | ||
| + | <WRAP column 15%> | ||
| + | {{ : | ||
| + | </ | ||
| + | |||
| + | <WRAP column 75%> | ||
| + | |||
| + | **Speaker**: | ||
| + | |||
| + | **Title**: What is and how to classify credibility of online health information? | ||
| + | |||
| + | **Abstract**: | ||
| + | Misinformation in online health content poses a significant threat to public health. Despite years of effort, both the medical and Internet research communities continue to struggle to develop reliable methods for its identification and classification. Manual assessment by domain experts is accurate but prohibitively expensive and difficult to scale. At the same time, many automated approaches rely on overly simplistic assumptions. For example, the vast majority of computational studies use binary TRUE/FALSE labels and employ unstandardized annotation protocols, making experimental results difficult to reproduce. | ||
| + | In this seminar, I will present key challenges in the detection and classification of medical misinformation, | ||
| + | |||
| + | **Biogram**: | ||
| + | Aleksandra Nabożny, PhD, specializes in the detection and analysis of medical disinformation, | ||
| + | </ | ||
| + | <WRAP clear></ | ||
| + | |||
| + | ==== 2026-03-05 ==== | ||
| + | <WRAP column 15%> | ||
| + | {{ : | ||
| + | </ | ||
| + | |||
| + | <WRAP column 75%> | ||
| + | |||
| + | **Speaker**: | ||
| + | |||
| + | **Title**: Large Language Models and Empirical Legal Studies. | ||
| + | |||
| + | **Abstract**: | ||
| + | The lecture examines the potential of large language models (LLMs) for empirical legal studies. Traditional empirical legal research has relied on labor-intensive annotation of legal texts to identify, e.g., legally relevant factors, thematic patterns, or other semantic categories. Recent experiments with LLMs demonstrate remarkable capabilities regarding zero- and few-shot semantic annotation of legal texts at levels approaching trained lawyers. However, significant challenges remain: model brittleness to prompt formatting, the need for subject-matter expert supervision, | ||
| + | |||
| + | **Biogram**: | ||
| + | Jaromir Savelka is a researcher associate in the Computer Science Department at Carnegie Mellon University. He is interested in the intersection of natural language processing and society. Jaromir' | ||
| + | </ | ||
| + | <WRAP clear></ | ||
| + | |||
| + | ==== 2026-01-29 ==== | ||
| + | <WRAP column 15%> | ||
| + | {{ : | ||
| + | </ | ||
| + | |||
| + | <WRAP column 75%> | ||
| + | |||
| + | **Speaker**: | ||
| + | |||
| + | **Title**: Calm-Data Generator: A Flexible Framework for Synthetic Dataset Creation Under Concept Drift. | ||
| + | |||
| + | **Abstract**: | ||
| + | Calm-Data Generator is designed to produce realistic synthetic datasets exhibiting different types of concept drift, enabling controlled evaluation of machine-learning models in dynamic environments. The framework implements multiple drift mechanisms, supports tabular data with customizable complexity, and provides a flexible API for defining data-generation functions, experimental scenarios, and temporal transitions. Its primary goal is to facilitate reproducible experimentation in concept drift research, allowing researchers to benchmark methods, analyze performance degradation, | ||
| + | |||
| + | **Biogram**: | ||
| + | Antonio Guillén-Teruel (a.guillenteruel@um.es) is a PhD Student graduated in Mathematics from the University of Murcia in 2020. In 2021 he got a Masters degree in Big Data from the same university and started his Ph.D studies in Informatics. The following year he got a Masters degree in Advanced Mathematics at the University of Murcia. His research focuses on imbalanced problems in Machine Learning (ML), including both in regression and classification problems, as well as the study of concept drift in medical domains for imbalanced datasets. | ||
| + | </ | ||
| + | <WRAP clear></ | ||
| + | |||
| + | ==== 2026-01-22 ==== | ||
| + | <WRAP column 15%> | ||
| + | {{ : | ||
| + | </ | ||
| + | |||
| + | <WRAP column 75%> | ||
| + | |||
| + | **Speaker**: | ||
| + | |||
| + | **Title**: Localizing the invisible: Graph Neural Networks for Biomedical Signals. | ||
| + | |||
| + | **Abstract**: | ||
| + | The respiratory system can be significantly affected by thoracic injuries, which may lead to complications such as lung dysfunction. Immediate diagnosis, along with the precise location of these injuries is crucial as it allows targeted medical interventions, | ||
| + | This talk introduces a wearable sensor network coupled with Temporal Graph Neural Networks (TGCNs) to detect and localize breathing abnormalities non-invasively and in real time. By modeling the thorax as a graph of 14 IMU-based sensor nodes, the system captures both spatial dependencies and temporal dynamics of chest movement. The proposed model pinpoints the exact location and severity of breathing irregularities, | ||
| + | The presentation will cover the underlying method, synthetic to real data strategy and the vision of a smart chest wearable system for continuous at home monitoring. Applications range from post-injury recovery and chronic disease management to early warning systems in preventive healthcare. | ||
| + | |||
| + | **Biogram**: | ||
| + | Zaidi is a PhD researcher at Otto-von-Guericke University Magdeburg, working at the Knowledge Management & Discovery Lab (KMD) under the supervision of Prof. Myra Spiliopoulou. His research focuses on graph neural networks, wearable sensor systems and feature importance for biomedical applications. | ||
| + | He has developed a temporal graph neural network framework to detect and localize respiratory abnormalities using chest-mounted IMU sensor networks. Beyond respiratory monitoring, his research spans spatio-temporal deep learning, health signal processing and interpretable machine learning methods aimed at bridging the gap between AI research and clinical practice. | ||
| + | |||
| + | </ | ||
| + | <WRAP clear></ | ||
| + | |||
| + | ==== 2026-01-15 ==== | ||
| + | <WRAP column 15%> | ||
| + | {{ : | ||
| + | </ | ||
| + | |||
| + | <WRAP column 75%> | ||
| + | |||
| + | **Speaker**: | ||
| + | |||
| + | **Title**: Knowledge Graphs and Polyvocality in Cultural Heritage and Beyond. | ||
| + | |||
| + | **Abstract**: | ||
| + | Over the past decade, Knowledge Graphs (KGs) have emerged as one of the most promising technologies for organizing and exploring cultural heritage data. Within this domain, several international standards have been developed to enable interoperable modelling of library, archive, and museum resources, such as CIDOC-CRM and LRMoo. Alongside these standards, new conceptual and methodological trends—such as polyvocality representation—are reshaping how cultural narratives and multiple perspectives can be expressed, connected, and contextualized within KGs. In this talk, I will share insights from my experience developing and applying KGs in two cultural heritage–related projects: CHExRISH and LKG. I will also discuss ongoing and future work on polyvocality in and beyond cultural heritage domains. | ||
| + | |||
| + | **Biogram**: | ||
| + | Luiz do Valle Miranda is a Ph.D. candidate at the Jagiellonian University in Technical Computer Science since 2023. He graduated in BA in Cognitive Science at the John Paul II Catholic University of Lublin (2018), received the title of MA in Philosophy from the University of Antwerp (2020) and has a Ph.D. in Philosophy from the Charles University in Prague (2024). | ||
| + | His research focuses on the development of knowledge-based systems, with particular emphasis on knowledge graphs. He is especially interested in the intersection of language model applications and explicitly represented knowledge, aiming to create intelligent technologies that promote human diversity and flourishing. | ||
| + | </ | ||
| + | <WRAP clear></ | ||
| + | |||
| + | ==== 2026-01-08 ==== | ||
| + | <WRAP column 15%> | ||
| + | {{ : | ||
| + | </ | ||
| + | |||
| + | <WRAP column 75%> | ||
| + | |||
| + | **Speaker**: | ||
| + | |||
| + | **Title**: Counterfactual Guidance for Transparent Hyperparameter Tuning. | ||
| + | |||
| + | **Abstract**: | ||
| + | This research explores how hyperparameter optimization can be made more transparent and interactive by incorporating human preferences into the search process. Instead of treating optimization as a black-box task, the approach uses counterfactual explanations to suggest alternative configurations under user-defined constraints. The method enables domain experts to better understand and guide model behavior, supporting more informed decision-making. | ||
| + | |||
| + | **Biogram**: | ||
| + | Sabri Manai is a PhD candidate in Technical Computer Science at Jagiellonian University in Kraków, where his research focuses on explainable AI and pattern detection in multimodal data. His work investigates how human feedback and domain knowledge can improve the transparency and reliability of AI systems. | ||
| + | He holds a Master’s degree in Software Systems Engineering from the Universitat Politècnica de València and a Bachelor’s degree in Computer Science from the South Mediterranean University in Tunis. Previously, he worked on AI-driven urban analytics within the Valencia Smart City project at Idrica and contributed to mobile development and cloud integration at Peaksource. | ||
| + | </ | ||
| + | <WRAP clear></ | ||
| ==== 2025-12-18 ==== | ==== 2025-12-18 ==== | ||
