| Both sides previous revisionPrevious revisionNext revision | Previous revision |
| aira:start [2024/06/24 12:30] – [2024-06-27] sbk | aira:start [2025/06/26 20:17] (current) – [Schedule Spring 2025] mtm |
|---|
| The program will be published at [[https://aira.geist.re]] in advance | The program will be published at [[https://aira.geist.re]] in advance |
| (a dedicated MS Teams group for announcements is available for those who are interested). | (a dedicated MS Teams group for announcements is available for those who are interested). |
| |
| Scientific secretary [[https://szymon.bobek.re|Szymon Bobek]] | |
| |
| Scientific coordination: [[https://gjn.re|Grzegorz J. Nalepa]] | Scientific coordination: [[https://gjn.re|Grzegorz J. Nalepa]] |
| | |
| | Scientific secretaries [[https://szymon.bobek.re|Szymon Bobek]], [[https://www.geist.re/pub:about_us:mtm|Maciej Mozolewski]], [[https://www.geist.re/pub:about_us:mzk|Maciej Szelążek]] |
| | |
| | |
| | ===== Schedule Spring 2025 ===== |
| | * **[RESEARCH TRACK] 2025.06.26**: Artur Miroszewski, PhD @ Jagiellonian University, [[#section20250626|Exploring Quantum Machine Learning through Earth Observation Case Studies.]] |
| | * Meeting link:[[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1750076384548?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7d|MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/Eb-PxGrj_L9Hka7sdu3TMY4BI2qKDq_5Bvl5hfUPyUwnNA?e=IFBlWx|View]] |
| | * Presentation slides: {{:aira:slides-artur-miroszewski-2025-06-26.pdf|Download}} |
| | |
| | * **[RESEARCH TRACK] 2025.06.05**: Elżbieta Sroka, PhD @ Łukasiewicz Research Network - EMAG Institute of Innovative Technologies, [[#section20250605| Interaction Design in Consideration of User Research and UX Specialist Perspectives.]] |
| | * Meeting link:[[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1748969648131?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7d|MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EXXr2Bww-WlGtD4kSo63kioBnlcBhajhN7zEXY22ok6huw?e=zPIl1S|View]] |
| | * Presentation slides: {{:aira:slides-elżbieta-sroka-2025-06-05.pdf|Download}} |
| | |
| | * **[PHD TRACK] 2025.05.29**: Bogdan Gulowaty, PhD @ Wrocław University of Science and Technology, [[#section20250529|Building transparent classification models.]] |
| | * Meeting link:[[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1748506829263?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7d|MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EZxjf0UnTCtChkK4Lk39qcIBDU9df_PNwuCe9D_2MpFn_w?e=GtoV3X|View]] |
| | * Presentation slides: {{:aira:slides-bogdan-gulowaty-2025-05-29.pdf|Download}} |
| | |
| | * **[PHD TRACK] 2025.05.22**: Jacek Karolczak, PhD Candidate @ Poznań University of Technology, [[#section20250522|Explainable AI: Moving from numbers to meaningful insights via prototype-based explanations.]] |
| | * Meeting link:[[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1747385677830?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7d|MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EYgrsIJ986RNpBAtv_11xLcBcDlCsAbrXYbUbMoTEMuSww?e=3g2tJQ|View]] |
| | * Presentation slides: {{:aira:slides-jacek karolczak-2025-05-22.pdf|Download}} |
| | |
| | * **[PHD TRACK] 2025.05.15**: Anastasiya Pechko, PhD Candidate @ Jagiellonian University, [[#section20250515|Adaptive Modular Housing Design for Crisis Situations.]] |
| | * Meeting link:[[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1746780764971?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7d|MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EbyiCvceYJdEnCVPvfgVlNsBVB5BDIsbHRpDRgc8_19rMA?e=wktReu|View]] |
| | * Presentation slides: {{:aira:slides-anastasiya-pechko-2025-05-15.pdf|Download}} |
| | |
| | * **[RESEARCH TRACK] 2025.05.08**: Gianluca Guglielmo, PhD @ Tilburg University, [[#section20250508|From Video Games to Real-life ”Games”: The Emergence of Real-life Expertise in (Serious) Video Games.]] |
| | * Meeting link:[[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1746526946846?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7d|MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EXI-yziJgQVEu8CHWewpTrABob_7ilL9hZeDNWU-OSbWWQ?e=nCNWWc|View]] |
| | * Presentation slides: {{:aira:slides-gianluca-guglielmo-2025-05-08.pdf|Download}} |
| | |
| | * **[PHD TRACK] 2025.04.24**: Natalia Wojak-Strzelecka, PhD Candidate @ Jagiellonian University, [[#section20250424|Enhancing concept drift detection, explanation and adaptation to changes in industrial data streams.]] |
| | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1744878780977?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7d|MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EWKQNaHF8SNGhAjftjZOkZcB07wrJAXFXi4-k7saiWYx6Q?e=Iq5cb0|View]] |
| | * Presentation slides: {{:aira:slides-natalia-wojak-strzelecka-2025-04-24.pdf|Download}} |
| | |
| | * **[PHD TRACK] 2025.04.03**: Dmytro Polishchuk, PhD Candidate @ Jagiellonian University, [[#section20250403|Automated GitHub Repository Quality Evaluation: A Metrics-Based Approach.]] |
| | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1743415239609?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7d|MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EWqf_EYlyY9PiBDp4s6kJDsBQo_W7RnsB9aP1WagKm0MRw?e=ROAopX|View]] |
| | * Presentation slides: {{:aira:slides-dmytro-polishchuk-2025-04-03.pdf|Download}} |
| | |
| | * **[PHD TRACK] 2025.03.27**: Jakub Jakubowski, PhD Candidate @ AGH University, [[#section20250327|Dry run thesis defense IN POLISH - Explainable Predictive Maintenance in Steel Rolling.]] |
| | * Meeting link: [[|The talk will be held stationary in room C-2-10]] |
| | * Recording: - (Dry run thesis defense) |
| | * Presentation slides: {{:aira:slides-jakub-jakubowski-20250327.pdf|Download}} |
| | |
| | * **[PHD TRACK] 2025.03.13**: Maciej Szelążek, PhD Candidate @ Jagiellonian University, [[#section20250313|Semantic Data Mining methods for decision support in smart manufacturing.]] |
| | * Meeting link: [[|offline mode]] |
| | * Recording: - (Dry run thesis defense) |
| | * Presentation slides: {{:aira:slides-maciej-szelążek-20250313.pdf|Download}} |
| | |
| | * **[RESEARCH TRACK] 2025.03.06**: Renata Włoch, Professor @ University of Warsaw, [[#section20250306|Does fear of automation motivate workers to reskill?]] |
| | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1740679635638?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7dv|MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EbvhD-bqRjdBt-xR1g59PCwBWucizrY-fXLASwAs17MVVw?e=CoomRp|View]] |
| | * Presentation slides: {{:aira:slides-renata-wloch-20250306.pdf|Download}} |
| | |
| | ===== Schedule Autumn 2024 ===== |
| | |
| | * **[PHD TRACK] 2025.01.30**: Mateusz Dobija, PhD Candidate @ Jagiellonian University, [[#section20250130|Accelerating training of Physics Informed Neural Network for 1D PDEs with Hierarchical Matrices]] |
| | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1737979076460?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7d|MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EWbuq7KTrU5FiOxvBuPGE-IBBf6J39JHbn2pexWaP0qNeg?e=KsoZgy|View]] |
| | * Presentation slides: {{:aira:slides-mateusz-dobija-20250130.pdf|Download}} |
| | |
| | * **[PHD GUEST TRACK] 2025.01.16**: Antonio Guillén-Teruel, PhD Candidate @ University of Murcia, [[#section20250116|Exploring SHAP Values in Imbalanced: Insights on Bias and Concept Drift]] |
| | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1736516251490?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7d|MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/Eax_Uqyzvq1GqQeb25IoZRcBsOb_qNU0pxzZWAjyGJwmjQ?e=ctGY93|View]] |
| | * Presentation slides: {{:aira:slides-Antonio-Guillen-20250116.pdf|Download}} |
| | |
| | * **[PHD GUEST TRACK] 2024.12.19**: Betül Bayrak, PhD Candidate @ Norwegian University of Science and Technology, [[#section20241219|Post-hoc XAI Methods: Counterfactuals and XCBR Applications]] |
| | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1734018722045?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7d|MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EXl4WJ32pppBmxIN1W433dMBAhVummmFfttPyOZ6qk2L2g?e=dGouZ0|View]] |
| | * Presentation slides: {{:aira:slides-20241219-Betül-Bayrak.pdf|Download}} |
| | |
| | * **[PHD TRACK] 2024.12.12**: Jan Ignatowicz, PhD Candidate @ Jagiellonian University, [[#section20241212|Knowledge Graphs for Digitized Manuscripts in Cultural Heritage Applications]] |
| | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1733481081680?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7d|MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/ET2WZunh3D5Hu1hDea6cxDoBm5SV1nZWZigJxf48h6hX0w?e=43N2aX|View]] |
| | * Presentation slides: {{:aira:slides-20241212-jan-ignatowicz.pdf|Download}} |
| | |
| | * **[PHD GUEST TRACK] 2024.12.05**: Ulvi Movsum-zada, PhD Candidate @ Jagiellonian University, [[#section20241205|VisTabNet: Adapting Vision Transformers for Tabular Data]] |
| | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1732960855409?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7d|MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EeEdC2iOLNdLgAB5YCqfTLUB_MOjmJZCA3_1n6BxcS3ayg?e=vTducN|View]] |
| | * Presentation slides: {{:aira:slides-20241205-Ulvi-Movsum-zada.pdf|Download}} |
| | |
| | * **[RESEARCH TRACK] 2024.11.28**: Michał Wierzchoń, Professor @ Jagiellonian University, [[#section20241128|Centre for Brain Research – aims, goals, case projects and available datasets]] |
| | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1732272338514?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7d|MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/ESzB-Y_DVG9OodchlbpLmywBYm_EywiY9LJnPtFTasLUoA?e=7QnVzg|View]] |
| | * Presentation slides: TDA |
| | |
| | * **[RESEARCH TRACK] 2024.11.21**: Agnieszka Ławrynowicz, Associate Professor @ Poznan University of Technology, [[#section20241121|How to Build Trustworthy Knowledge Graphs?]] |
| | * Meeting link: [[https://meet.google.com/iwh-rejz-tdr|MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EZJsuG40DehFkCbFySSjkWcBxnm452xn5eo7qdVzBdnVOA?e=lX5pde|View]] |
| | * Presentation slides: TDA |
| | |
| | * **[RESEARCH TRACK] 2024.11.14**: Dorota Głowacka, Professor @ University of Helsinki, [[#section20241114|User-centric Design and Evaluation of Exploratory Search and Recommender Systems]] |
| | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1731059980619?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7d |MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EbZ78VBLZe1JoXmI1fAoWHgBep-ffncV7EMxxxg8btH78g?e=MIc7x4|View]] |
| | * Presentation slides: {{:aira:slides-20241114-dorota-glowacka-compressed.pdf |Download}} |
| | * |
| | * **[PHD GUEST TRACK] 2024.11.07**: Weronika Hryniewska-Guzik, [[#section20241107|NormEnsembleXAI: Unveiling the Strengths and Weaknesses of XAI Ensemble Techniques]] |
| | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1729783311580?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7d |MS Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/Ef0Yqg9j8CtDkw215CaFVmABbWmQCeUf_Npmksnzo0mRtQ?e=dGLK0M|View]] |
| | * Presentation slides: {{:aira:slides-20241107-weronika-hryniewska-guzik.pdf |Download}} |
| | |
| | * **[RESEARCH TRACK] 2024.10.24**: Lucjan Janowski, [[#section20241024|Quality of Experience - Quality for Telecommunication]] |
| | * Meeting link: [[https://teams.microsoft.com/l/message/19:JF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1@thread.tacv2/1729175414324?tenantId=eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb&groupId=8507cdef-d08b-4352-a3af-fa1b61acec4c&parentMessageId=1729175414324&teamName=(AIRA)%20Artificial%20Intelligence%20in%20Research%20and%20Applications%20Seminar&channelName=General&createdTime=1729175414324|MT Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EbFngP17Y8dMiiSWLcW0hLEBE7JkqX6lVil3wbasZpdzcw?e=D6i6cT|View]] |
| | * Presentation slides: {{:aira:slides-20241024-Lucjan-Janowski.pdf |Download}} |
| | |
| | * **[DOCTORAL TRACK] 2024.10.17**: Luiz do Valle Miranda, [[#section20241017|Critical Early-Stage Decisions in Linked Data Projects for Cultural Heritage: Challenges, Choices, and Guidelines]] |
| | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1728683060737?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22a39bd3b1-7b43-47ab-b5ec-29d9ab0ccbb2%22%7d|MT Teams]] |
| | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EUXXf_L4sONBoEmRBqct5qcBXraen1Z0L5DDGpDp0NBwDg?e=ulnNVP|View]] (if you are not UJ employee, ask Maciej Mozolewski or Maciej Szelążek for access) |
| | * Presentation slides: {{:aira:slides-20241017-luiz-do-valle-miranda-v2.pdf |Download}} |
| |
| ===== Schedule Summer 2024 ===== | ===== Schedule Summer 2024 ===== |
| * **[RESEARCH TRACK] 2024.06.27**: José Palma,Juan Botía, Antonio Guillén-Teruel [[#20240627| Context-Aware learning models: CALM-Project]] | * **[RESEARCH TRACK] 2024.06.27**: José Palma,Juan Botía, Antonio Guillén-Teruel [[#20240627| Context-Aware learning models: CALM-Project and Concept Drift in Imbalanced Problems]] |
| * Meeting link: [[|MS Teams]] | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1719232789438?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22c96e4eee-96f3-4b0f-88ef-fe65310f5f55%22%7d|MS Teams]] |
| * Recording: [[|View]] (if you are not UJ employee, ask Szymon Bobek for access) | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EW98gKjTKOpDmWvqQrrWoIMBN5tJHbO_BtEksP0JGsrAvA?e=lDLkSh|View]] (if you are not UJ employee, ask Szymon Bobek for access) |
| * Presentation slides: {{ |Download}} | * Presentation slides: {{:aira:slides-jose-palma-20240617.pdf |Download I}} {{ :aira:slides-antonio-guillen-20240627.pdf |Download II}} |
| * **[RESEARCH TRACK] 2024.06.06**: Żaneta Kubic [[#20240606| European Cultural Heritage in Virtual Worlds – why and how : introduction to the IMPULSE project]] | * **[RESEARCH TRACK] 2024.06.06**: Żaneta Kubic [[#20240606| European Cultural Heritage in Virtual Worlds – why and how : introduction to the IMPULSE project]] |
| * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1717407321901?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22c96e4eee-96f3-4b0f-88ef-fe65310f5f55%22%7d|MS Teams]] | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1717407321901?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22c96e4eee-96f3-4b0f-88ef-fe65310f5f55%22%7d|MS Teams]] |
| * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EfkSlUmXWbRGvq0xS6bBZkcBrrRI2bumOitE0ou0VBKC2w?nav=eyJyZWZlcnJhbEluZm8iOnsicmVmZXJyYWxBcHAiOiJTdHJlYW1XZWJBcHAiLCJyZWZlcnJhbFZpZXciOiJTaGFyZURpYWxvZy1MaW5rIiwicmVmZXJyYWxBcHBQbGF0Zm9ybSI6IldlYiIsInJlZmVycmFsTW9kZSI6InZpZXcifX0%3D&e=3EfKGG|View]] (if you are not UJ employee, ask Szymon Bobek for access) | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EfkSlUmXWbRGvq0xS6bBZkcBrrRI2bumOitE0ou0VBKC2w?nav=eyJyZWZlcnJhbEluZm8iOnsicmVmZXJyYWxBcHAiOiJTdHJlYW1XZWJBcHAiLCJyZWZlcnJhbFZpZXciOiJTaGFyZURpYWxvZy1MaW5rIiwicmVmZXJyYWxBcHBQbGF0Zm9ybSI6IldlYiIsInJlZmVycmFsTW9kZSI6InZpZXcifX0%3D&e=3EfKGG|View]] (if you are not UJ employee, ask Szymon Bobek for access) |
| * Presentation slides: {{ |Download}} | * Presentation slides: {{ :aira:slides-zaneta-kubic-20240606.pdf |Download}} |
| * **[DOCTORAL TRACK] 2024.05.16**: Mateusz Bułat [[#20240516| Image analysis for specialised therapy support]] | * **[DOCTORAL TRACK] 2024.05.16**: Mateusz Bułat [[#20240516| Image analysis for specialised therapy support]] |
| * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1715342946914?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22c96e4eee-96f3-4b0f-88ef-fe65310f5f55%22%7d|MS Teams]] | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1715342946914?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22c96e4eee-96f3-4b0f-88ef-fe65310f5f55%22%7d|MS Teams]] |
| * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1713853200005?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22c96e4eee-96f3-4b0f-88ef-fe65310f5f55%22%7d|MS Teams]] | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1713853200005?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22c96e4eee-96f3-4b0f-88ef-fe65310f5f55%22%7d|MS Teams]] |
| * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EXQKuYkW3VFMr94bDgX3kR8BEDi5SYRLog-xkyoNA0ONJA?nav=eyJyZWZlcnJhbEluZm8iOnsicmVmZXJyYWxBcHAiOiJTdHJlYW1XZWJBcHAiLCJyZWZlcnJhbFZpZXciOiJTaGFyZURpYWxvZy1MaW5rIiwicmVmZXJyYWxBcHBQbGF0Zm9ybSI6IldlYiIsInJlZmVycmFsTW9kZSI6InZpZXcifX0%3D&e=T6Cnc7|View]] (if you are not UJ employee, ask Szymon Bobek for access) | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EXQKuYkW3VFMr94bDgX3kR8BEDi5SYRLog-xkyoNA0ONJA?nav=eyJyZWZlcnJhbEluZm8iOnsicmVmZXJyYWxBcHAiOiJTdHJlYW1XZWJBcHAiLCJyZWZlcnJhbFZpZXciOiJTaGFyZURpYWxvZy1MaW5rIiwicmVmZXJyYWxBcHBQbGF0Zm9ybSI6IldlYiIsInJlZmVycmFsTW9kZSI6InZpZXcifX0%3D&e=T6Cnc7|View]] (if you are not UJ employee, ask Szymon Bobek for access) |
| * Presentation slides: {{ |Download}} | * Presentation slides: {{ :aira:slides-bartek-malkus-20240425.pdf|Download}} |
| * **[DOCTORAL TRACK] 2024.04.18** | * **[DOCTORAL TRACK] 2024.04.18** |
| * Farnoud Ghasemi [[#20240418| Performance Optimization of the Platforms in Two-sided Mobility Market]] and | * Farnoud Ghasemi [[#20240418| Performance Optimization of the Platforms in Two-sided Mobility Market]] and |
| * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1687248545347?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22c96e4eee-96f3-4b0f-88ef-fe65310f5f55%22%7d|MS Teams]] | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1687248545347?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22c96e4eee-96f3-4b0f-88ef-fe65310f5f55%22%7d|MS Teams]] |
| * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EWq81Fq6gW9Pq49RKHA99_0BAzdBFtCX5SKzUM0LoP-ULQ?e=1NzMwJ&nav=eyJyZWZlcnJhbEluZm8iOnsicmVmZXJyYWxBcHAiOiJTdHJlYW1XZWJBcHAiLCJyZWZlcnJhbFZpZXciOiJTaGFyZURpYWxvZyIsInJlZmVycmFsQXBwUGxhdGZvcm0iOiJXZWIiLCJyZWZlcnJhbE1vZGUiOiJ2aWV3In19|View]] (if you are not UJ employee, ask Szymon Bobek for access) | * Recording: [[https://ujchmura.sharepoint.com/:v:/t/Section_495645_1/EWq81Fq6gW9Pq49RKHA99_0BAzdBFtCX5SKzUM0LoP-ULQ?e=1NzMwJ&nav=eyJyZWZlcnJhbEluZm8iOnsicmVmZXJyYWxBcHAiOiJTdHJlYW1XZWJBcHAiLCJyZWZlcnJhbFZpZXciOiJTaGFyZURpYWxvZyIsInJlZmVycmFsQXBwUGxhdGZvcm0iOiJXZWIiLCJyZWZlcnJhbE1vZGUiOiJ2aWV3In19|View]] (if you are not UJ employee, ask Szymon Bobek for access) |
| * Presentation slides: TBA | * Presentation slides: {{ :aira:slides-20230629-sepideh-pashami.pdf |Download}} |
| * **[TUTORIAL] 2023.06.26**: Sepideh Pashami [[#20230626| Causal Inference and its Connection to Machine Learning]] (//Project CHIST-ERA XPM//) | * **[TUTORIAL] 2023.06.26**: Sepideh Pashami [[#20230626| Causal Inference and its Connection to Machine Learning]] (//Project CHIST-ERA XPM//) |
| * **LOCATION: A2-02** (This meeting will be in hybrid mode: on-site and online) | * **LOCATION: A2-02** (This meeting will be in hybrid mode: on-site and online) |
| * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1687174755583?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22c96e4eee-96f3-4b0f-88ef-fe65310f5f55%22%7d|MS Teams]] | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1687174755583?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22c96e4eee-96f3-4b0f-88ef-fe65310f5f55%22%7d|MS Teams]] |
| * Presentation slides: TBA | * Presentation slides: {{ :aira:slides-20230626-sepideh-pashami.pdf |Download}} |
| * **[DOCTORAL TRACK] 2023.06.22**: Michał Kuk [[#20230622| Time-Series Complexity into Understandable Prototypes: A Generic Approach to Machine Learning Explanations in Industrial Processes]] (//Project CHIST-ERA XPM//) | * **[DOCTORAL TRACK] 2023.06.22**: Michał Kuk [[#20230622| Time-Series Complexity into Understandable Prototypes: A Generic Approach to Machine Learning Explanations in Industrial Processes]] (//Project CHIST-ERA XPM//) |
| * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1687177218502?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22c96e4eee-96f3-4b0f-88ef-fe65310f5f55%22%7d|MS Teams]] | * Meeting link: [[https://teams.microsoft.com/l/meetup-join/19%3aJF1L0935A7eV6s8R9brG5MMYONrqy4XmxPSYLeRMCGM1%40thread.tacv2/1687177218502?context=%7b%22Tid%22%3a%22eb0e26eb-bfbe-47d2-9e90-ebd2426dbceb%22%2c%22Oid%22%3a%22c96e4eee-96f3-4b0f-88ef-fe65310f5f55%22%7d|MS Teams]] |
| |
| ===== Presentation details ===== | ===== Presentation details ===== |
| | |
| | |
| | |
| | ==== 2025-06-26 ==== |
| | <WRAP column 15%> |
| | {{ :aira:artur-miroszewski-foto.jpeg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Artur Miroszewski, PhD @ Jagiellonian University |
| | |
| | **Title**: Exploring Quantum Machine Learning through Earth Observation Case Studies. |
| | |
| | **Abstract**: |
| | The analysis of satellite images has attracted significant research interest due to its numerous applications and unparalleled scalability in Earth Observation (EO). Although artificial intelligence algorithms for EO emerge at a steady pace, the community still needs to address many practical challenges that are concerned with such highly dimensional and unprecedentedly large volumes of image data. Quantum Machine Learning (QML) is a promising research avenue here. Despite the growing interest and funding in the field, current results remain inconclusive, shifting focus toward understanding the strengths and limitations of QML rather than solely expanding applications of such methods. This seminar addresses a critical gap by consolidating advancements in QML, with a special focus put on quantum kernel methods - which already proved their value in EO - to evaluate their role in advancing state-of-the-art EO solutions and exploring the potential quantum advantage via identifying their benefits and shortcomings in an unbiased and thorough way. |
| | |
| | **Biogram**: |
| | Artur Miroszewski received the Ph.D. degree in theoretical physics from the National Centre for Nuclear Research, Otwock, Poland, in 2021. He is a Postdoctoral Researcher with the Jagiellonian University, Kraków, Poland. |
| | He is involved in European Space Agency projects exploring the potential of quantum machine learning for satellite data analysis and serves as a quantum computing lecturer at the IEEE GRSS HDCRS summer schools. He is a co-chair of the QUEST IEEE GRSS Technical Committee. |
| | His main research interest include the study and applications of quantum kernel methods. |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | ==== 2025-06-05 ==== |
| | <WRAP column 15%> |
| | {{ :aira:elzbieta-sroka-foto.jpg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Elżbieta Sroka, PhD @ Łukasiewicz Research Network - EMAG Institute of Innovative Technologies |
| | |
| | **Title**: Interaction Design in Consideration of User Research and UX Specialist Perspectives: Artificial intelligence in UX Design, users experience in digital library and the needs of digital humanities researchers, reception of signed avatars by Deaf users. |
| | |
| | **Abstract**: |
| | The presentation presents the results of research in the field of Human-Computer Interaction (HCI), with a special focus on the needs of different user groups and the application of artificial intelligence (AI) tools in both the design and research processes. The areas discussed include: the use of AI in UX design and designers' perspectives; the experiences of digital library users, especially in the context of orientation in the field, navigation, and information retrieval; the needs of digital collection researchers and their expectations regarding AI-supported solutions; and the accessibility of research tools for Deaf people and the reception of sign avatars communicating in Polish Sign Language (PJM) in the Deaf community. |
| | |
| | **Biogram**: |
| | Elżbieta Sroka, PhD, certified UX designer, Senior Specialist at the Łukasiewicz Research Network – Institute of Innovative Technologies EMAG in Katowice, Poland. |
| | She obtained her doctoral degree in 2018 from the University of Silesia in Katowice, based on a dissertation focused on the digitization of social life documents in Polish digital libraries. |
| | Her research interests include research users information behavior, user experience (UX) design, and digital humanities, as well as applications of artificial intelligence—particularly in the context of human–AI interaction, the impact of AI on UX, and the use of AI in the study of digital collections. She also conducts research in the areas of digital accessibility, information management and information retrieval. |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | |
| | ==== 2025-05-29 ==== |
| | <WRAP column 15%> |
| | {{ :aira:Bogdan_Gulowaty_foto.jpg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Bogdan Gulowaty, PhD @ Wrocław University of Science and Technology |
| | |
| | **Title**: Building transparent classification models. |
| | |
| | **Abstract**: |
| | As AI and ML technologies advance and become deeply integrated into daily life, the need for transparent and interpretable models grows increasingly urgent. AI systems must be understandable, trustworthy, and ethical, especially in critical healthcare, finance, and legal sectors. The rise of XAI seeks to address these challenges by providing explanations of model decisions, making AI systems more transparent. However, much of the progress in XAI has focused on deep neural networks, leaving other complex models like ensemble methods needing to be explored more regarding their interpretability. |
| | |
| | The thesis aims to fill that gap by developing novel methods to improve the transparency and interpretability of ensemble classifiers while ensuring these models maintain competitive predictive performance and build inherently transparent models. The central hypothesis of the thesis is that it is possible to construct such transparent or explainable models that perform as well as black-box models in a wide range of classification tasks. The work focuses on three primary methods designed to either explain or replace complex models with transparent alternatives: NOTE, optimal-centroids and quad-split algorithms. |
| | |
| | **Biogram**: |
| | Finished his PhD at Wroclaw University of Technology in 2025. Full time Software Engineer. Enthusiast of mountaineering, various sports activities and motorcycling. Born in Bolesławiec, currently lives in Wrocław. |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | ==== 2025-05-22 ==== |
| | <WRAP column 15%> |
| | {{ :aira:Jacek_Karolczak_foto.jpg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Jacek Karolczak, PhD Candidate @ Poznań University of Technology |
| | |
| | **Title**: Explainable AI: Moving from numbers to meaningful insights via prototype-based explanations. |
| | |
| | **Abstract**: |
| | The increasing complexity of machine learning models has heightened the demand for their explainability. Most existing work in explainable artificial intelligence (xAI) focuses on techniques like pixel attribution or feature importance, especially Shapley values. However, these types of explanations are often criticized for being hard to interpret—not just for laypersons, but even for machine learning experts. In contrast, the XAI 2.0 manifesto advocates for concept-based explanations, such as prototypes - representative instances. This talk will introduce the problem and survey existing approaches, with a focus on recent developments in prototype-based explainability. It will also present the author’s own work on prototype-based concept drift detection, which maintains intrinsic interpretability. Finally, open challenges and directions for future work in prototype-based explainability will be discussed. |
| | |
| | **Biogram**: |
| | Jacek Karolczak is a PhD student at Poznan University of Technology, where he also earned BSc and MSc in Artificial Intelligence. His research focuses on improving the interpretability of machine learning models, particularly in dynamic environments where data continuously evolves. He believes the world of explainable AI (xAI) is shifting beyond feature-importance explanations, embracing high-level concept-based interpretations that better align with the language and reasoning of end users. |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | |
| | ==== 2025-05-15 ==== |
| | <WRAP column 15%> |
| | {{ :aira:Anastasiya_Pechko_foto.png?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Anastasiya Pechko, PhD Candidate @ Jagiellonian University |
| | |
| | **Title**: Adaptive Modular Housing Design for Crisis Situations. |
| | |
| | **Abstract**: |
| | Today, the world is facing many challenges, and one of the most pressing is the humanitarian crisis caused by the war in Ukraine. A large number of people have been displaced and left without homes. That's why the need for scalable, adaptable housing solutions is highly relevant. This talk presents a new way of using computational tools—specifically, an adapted version of the Wave Function Collapse (WFC) algorithm—to design modular housing estates. The applied heuristics allow the solutions o meet specific project requirements, generating various modular settlement designs that consider functionality and social aspects. The talk will include examples of generated modular arrangements, highlighting the potential of this approach in real-world applications. |
| | |
| | **Biogram**: |
| | Anastasiya Pechko is a first-year PhD student in Technical Computer Science with a master's degree in computer game science (2023). She is a member of the Neu3D research group led by Dr. Przemysław Spurek and is also involved in the project "Effective Rendering of 3D Objects Using Gaussian Splatting in an Augmented Reality Environment" under the FIRST TEAM FENG programme of the Foundation for Polish Science. Her research focuses on computer-aided design and neural rendering—particularly Gaussian Splatting. In addition, Anastasiya has a deep interest in the synergy between art and algorithms, with a particular fascination for complex, emergent behaviors in digital systems. |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | ==== 2025-05-08 ==== |
| | <WRAP column 15%> |
| | {{ :aira:gianluca-guglielmo-foto.png?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Gianluca Guglielmo, PhD @ Tilburg University |
| | |
| | **Title**: From Video Games to Real-life ”Games”: The Emergence of Real-life Expertise in (Serious) Video Games. |
| | |
| | **Abstract**: |
| | Can real-life behaviors carry over into digital environments like serious and video games? Do we make decisions and respond to events in simulated settings the same way we do in the real world? These are the underlying questions that Gianluca Guglielmo will explore in this talk. Gianluca Guglielmo is a researcher who recently obtained his PhD in Cognitive Science and Artificial Intelligence at Tilburg University (Netherlands). In this seminar presentation, Gianluca will present some results of his PhD project, which was conducted in collaboration with Port of Rotterdam. During this project, he focused on using serious games to address some future challenges that the Port of Rotterdam will face in the future. Such challenges include the green transition, in line with the European Green Deal goals, as well as identifying new potential employees. These objectives are based on the idea that serious games, like video games, over safe simulated environments where specific skills can emerge and be developed. Throughout his PhD, Gianluca applied methods drawn from cognitive science, data science, and game studies. He combined non-invasive techniques to track physiological responses with machine learning algorithms to evaluate the effectiveness of a fordable and scalable approaches that can be used both in research with limited resources and in business contexts. The methods proposed may be applied not only to serious and video games but also to other screen-based tasks across different domains. The results presented in this seminar presentation will provide evidence that skills acquired in real life also manifest in serious games and that this transferability overs a solid foundation for using serious games to simulate future developments within companies and to help identify |
| | the experts of tomorrow." |
| | |
| | |
| | |
| | **Biogram**: |
| | Gianluca Guglielmo is an Italian researcher born and raised in Milan. He recently earned his PhD in Cognitive Science and Artificial Intelligence at Tilburg University. His PhD project, conducted in collaboration with the Port of Rotterdam, focused on decision-making, expertise, and the use of serious games. The project aimed to demonstrate how serious games can be powerful tools not only for communicating innovative business concepts to stakeholders but also for identifying and selecting new employees. |
| | Academically, Gianluca has a diverse and interdisciplinary background. He began with a Bachelor's degree in Philosophy at Università Statale di Milano, where he specialized in |
| | philosophy of science writing a thesis focused on game theory and evolutionary game theory. His interest in philosophy reflects his strong belief that the way we define a problem |
| | fundamentally shapes how we investigate it and the research outcomes we obtain. He then pursued a Master's in Cognitive Science and Neuroscience through a joint program at |
| | Università Statale di Milano and Maastricht University, majoring in psychopharmacology. During this time, he joined a research team at San Paolo Hospital in Milan, working on the |
| | clinical use of ketamine as a fast-action antidepressant for patients with major depressive disorder. His contributions helped produce the first Italian case series on this treatment, which also formed the basis of his master’s thesis. Before starting his PhD, Gianluca completed a second Master’s in Cognitive Science and |
| | Artificial Intelligence. During this period, he worked as a student assistant on a project employing video games to study moral decision-making. This research culminated in a |
| | conference paper titled "The Temperature of Morality: A Behavioral Study on the Efect of Moral Decisions on Facial Thermal Variations in Video Games." |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | ==== 2025-04-24 ==== |
| | <WRAP column 15%> |
| | {{ :aira:natalia-wojak-strzelecka-foto.jpg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Natalia Wojak-Strzelecka, PhD Candidate @ Jagiellonian University |
| | |
| | **Title**: Automated GitHub Repository Quality Evaluation: A Metrics-Based Approach. |
| | |
| | **Abstract**: |
| | In this seminar, we will explore three complementary approaches to handling evolving data in industrial environments. We'll discuss methods for detecting and adapting to domain shifts in data streams, distinguishing between real system failures and normal process changes, and using explainable AI to better understand and interpret concept drift. The presented work combines domain adaptation, drift detection, and XAI to improve the robustness and transparency of machine learning models in real-time settings like manufacturing and healthcare. |
| | |
| | **Biogram**: |
| | Natalia has received Bachelor's (2020) and Master's (2022) degrees in Mathematics from Silesia Univerity of Technology, Faculty of Applied Mathematics. Her career path is deeply rooted in the industry, she started as a data scientist working on vibration signals for predictive maintenance applications and continuing as a modelling specialist at ArcelorMittal, where she develops and implements models for production optimization and image processing. Currently, as a PhD candidate, she is working on advanced domain adaptation techniques for industrial data stream applications and explainable anomaly detection. |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | ==== 2025-04-03 ==== |
| | <WRAP column 15%> |
| | {{ :aira:dmytro-polishchuk-foto.png?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Dmytro Polishchuk, PhD Candidate @ Jagiellonian University |
| | |
| | **Title**: Automated GitHub Repository Quality Evaluation: A Metrics-Based Approach. |
| | |
| | **Abstract**: |
| | This research aims to develop an automated system for evaluating the quality of GitHub repositories using a pre-established quality model. The system will assess repositories based on a variety of quality metrics, such as the commit history and its associated metadata (e.g., size, date, description), code coverage, pull request review time, issue resolution time, number of open issues, code churn, and code complexity. Additional metrics may also be considered, potentially including those defined by standards like ISO/IEC 25010:2011. |
| | |
| | **Biogram**: |
| | Software engineer with a background in the telecom domain, specializing in network management systems and Software Defined Networking (SDN). Experienced in performance engineering, including byte code instrumentation, and has worked across various technologies, including IoT. Previously contributed to major companies like Ericsson, Cisco Systems, and Playtech. |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | ==== 2025-03-27 ==== |
| | <WRAP column 15%> |
| | {{ :aira:jakub-jakubowski-foto.jpg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Jakub Jakubowski, PhD Candidate @ AGH University |
| | |
| | **Title**: Dry run thesis defense IN POLISH - Explainable Predictive Maintenance in Steel Rolling. |
| | |
| | **Abstract**: |
| | In recent years, there has been a growing interest in Industry 4.0, which seeks to integrate digital technologies into manufacturing processes. One key area that stands to benefit from these advancements is maintenance of the equipment. Artificial Intelligence (AI) can play a crucial role in developing Predictive Maintenance (PdM) solutions, which aim to reduce downtime, lower maintenance costs, and enhance safety in manufacturing environments. These technologies are particularly valuable in steel manufacturing, a resource-intensive and economically critical industry. However, a significant challenge in deploying AI-based PdM solutions in production is the lack of transparency, as many AI methods function as "black boxes." Without a clear understanding of the AI’s decision-making process, operators and engineers may struggle to take appropriate corrective actions. The research conducted by the presenter focused on the development and implementation of Explainable Artificial Intelligence (XAI) techniques in the steel rolling process, a critical step in steel production. This presentation summarizes the PhD thesis, which addresses the discussed problems. |
| | |
| | **Biogram**: |
| | Jakub Jakubowski earned his Bachelor's degree in Energy Engineering from AGH University of Science and Technology in 2016, followed by a Master's degree in 2017 from the Faculty of Fuels and Energy. Since 2018, he has been working at ArcelorMittal, the world's largest steel producer, as a modeling specialist and data scientist. His responsibilities include developing and implementing mathematical models to optimize manufacturing processes. Additionally, he assists engineers in analyzing large-scale industrial data and developing business intelligence tools. In 2020, he completed postgraduate studies in Data Science at AGH UST's Faculty of Computer Science, Electronics, and Telecommunications. That same year, he became a PhD candidate at AGH UST, participating in the Implementation Doctorate Programme, which integrates academic research with industry work. His PhD thesis has been completed and is currently under review. His primary research interest is the application of AI techniques in industrial settings, particularly in predictive maintenance solutions. |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | ==== 2025-03-13 ==== |
| | <WRAP column 15%> |
| | {{ :aira:maciej-szelazek.jpg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Maciej Szelążek, PhD Candidate @ Jagiellonian University |
| | |
| | **Title**: Semantic Data Mining methods for decision support in smart manufacturing. |
| | |
| | **Abstract**: |
| | This thesis focuses on integrating Machine Learning (ML) and Explainable Artificial Intelligence (XAI) techniques in a suitable way for their effective use in the decision–making process in the industrial Quality Management (QM) field. Continuous improvement in product and process quality is essential for manufacturers to stay competitive, and ML is increasingly being used for tasks like production planning or advanced control of production line devices. However, the complexity of ML models often makes their results difficult for non–expert users to interpret, limiting their practical applications. To overcome this challenge, we introduces Semantic Data Mining (SDM) concept that refers, in this thesis, to utilisation of ML models along with explainability methods in industrial quality control tasks. We considered SDM approach to detect patterns and extracting inferences from data that are both meaningful and interpretable by end users, providing easy to understand and act upon insights. This ensures that ML–generated results can be effectively used in decision–making processes. |
| | The thesis explores three main research challenges: integrating XAI insights into QM systems, developing visualization methods that fuse ML outputs with existing quality control tools, and applying SDM to enhance decision–making in industrial contexts. By proposing a new SDM component for QM, the research aims to improve the usability and impact of ML models in smart manufacturing. The study also focuses on ensuring that these new approaches can be seamlessly integrated with current industry standards and practices, making them applicable in a wide range of industrial settings. Through the development of these methods, the research contributes to the advancement of smart manufacturing, enabling more accurate and actionable quality control decisions. Presented thesis consists o fa series of publications that explore various aspects of the SDM approach for industrial quality control. Collectively, they offer a comprehensive analysis of the QM field and advance both the theoretical understanding and practical aspects of incorporating ML models to facilitate decision–making for smart manufacturing. |
| | |
| | **Biogram**: |
| | Maciej Szelążek is a member of the GEIST research team led by Prof. Gregory J. Nalepa (Jagiellonian University). Conducting research in the areas of machine learning, modeling, statistics and Semantic Data Mining (SDM) methods. International research collaboration in projects related to predictive maintenance, quality optimization, and state-of-the-art methods for unsupervised data evaluation using explainable AI (XAI). |
| | He worked as a data analyst in the Statistical Process Control (SPC) Office of Arcelor Mittal Poland. |
| | He participated in the creation and development of an analytical system, as well as multidimensional big data analysis related to bottleneck finding, logistics, cost optimization and process variability reduction. |
| | In the role of Data Scientist at Comarch Sp. z o.o., he carried out a project related to predicting customer behavior to increase the effectiveness of marketing campaigns. Responsibilities included the design, execution and implementation in a production environment of a procedure based on a machine learning model. |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | |
| | ==== 2025-03-06 ==== |
| | <WRAP column 15%> |
| | {{ :aira:renata-wloch-foto.png?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Renata Włoch, Professor @ University of Warsaw |
| | |
| | **Title**: Does fear of automation motivate workers to reskill? |
| | |
| | **Abstract**: |
| | As AI-driven automation rapidly reshapes workplaces, concerns about job displacement have grown. However, beyond the widely discussed fear of automation, how are workers actually responding? While mainstream discussions present reskilling as a simple solution to labor market shifts, our research takes a critical approach, exploring how both structural and individual factors influence workers’ training motivations. |
| | |
| | Drawing on survey data from six European countries (Austria, Czechia, Germany, Hungary, Poland, and Slovakia), we explore how exposure to technology, labor market vulnerability, and perceptions of control shape automation anxiety. Our findings show that workers in highly routinized and substitutable occupations experience heightened fear of automation, yet paradoxically exhibit lower willingness to engage in training. While exposure to technology intensifies automation fears, it does not necessarily translate into proactive skill acquisition—particularly for lower-income and lower-educated workers who lack institutional support for reskilling. Instead, training motivation is highest among workers in AI-augmented roles, where upskilling aligns with career advancement rather than job survival. |
| | |
| | This research challenges the dominant rhetoric of lifelong learning, which assumes that all workers can continuously retrain to remain employable. By situating reskilling within the broader sociotechnical landscape, we highlight how labor market structures, employer-supported training schemes, and national policies shape access to learning opportunities. Rather than framing automation as an inevitable force requiring individual adaptation, we call for policy approaches that prioritize structural interventions, ensuring equitable access to training and fostering worker agency in shaping the future of work. |
| | |
| | **Biogram**: |
| | Renata Włoch is Professor at the Faculty of Sociology, University of Warsaw, where she leads the Department of Digital Sociology. She serves as the Scientific Director of the Digital Economy Lab (DELab), an interdisciplinary center of research excellence focused on the societal and economic impacts of digitalization. With a strong background in qualitative and applied research, she has advised public institutions and produced research reports for businesses and NGOs. Since 2014, her work has centered on digital transformation, culminating in her co-authorship of The Economics of Digital Transformation (Routledge, 2021, with K. Śledziewska), which examines its effects on labor markets and business practices. She is currently conducting research on the development of workers’ digital skills as part of two international consortia within the Horizon INAiR project (focused on the retail sector) and the Erasmus USMED project (focused on the accommodation and food sector). |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | ==== 2025-01-30 ==== |
| | <WRAP column 15%> |
| | {{ :aira:mateusz-dobija-foto.jpg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Mateusz Dobija, PhD Candidate @ Jagiellonian University |
| | |
| | **Title**: Accelerating training of Physics Informed Neural Network for 1D PDEs with Hierarchical Matrices |
| | |
| | **Abstract**: |
| | We consider a training of Physics Informed Neural Networks with fully connected neural networks for approximation of solutions of one-dimensional advection-diffusion problem. In this context, the neural network is interpreted as a non-linear function of one spatial variable, approximating the solution scalar field, namely y = PINN(x) = Anσ(An−1...A2σ(A1 + b1) + b2) + ... + bn−1) + bn. In the standard PINN approach, the Ai denotes dense matrices, bi denotes bias vectors, and σ is the non-linear activation function (sigmoid |
| | in our case). In our paper, we consider a case when Ai are hierarchical matrices Ai = Hi. We assume a structure of our hierarchical matrices approximating the structure of finite difference matrices employed to solve analogous PDEs. In this sense, we propose a hierarchical neural network for training and approximation of PDEs using the PINN method. We verify our method on the example of a one-dimensional advection-diffusion |
| | problem. |
| | |
| | **Biogram**: |
| | Mateusz Dobija (mateusz.dobija@doctoral.uj.edu.pl) is a PhD candidate at the Jagiellonian University in the field of Technical Computer Science since 2021. He received BCs degree in Computer Science at the Faculty of Physics, Astronomy and Applied Computer Science of the Jagiellonian University in 2018, and MSc degree from Applied Computer Science at the Faculty of Physics, Astronomy and Applied Computer of the Jagiellonian University in 2020. He is interested in speeding up the MES/rIGA computations with the usage of hierarchical matrices. Lately working on the Physics Informed Neural Networks. |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | ==== 2025-01-16 ==== |
| | <WRAP column 15%> |
| | {{ :aira:antonio-teruel-foto.jpeg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Antonio Guillén-Teruel, PhD Candidate @ University of Murcia |
| | |
| | **Title**: Exploring SHAP Values in Imbalanced: Insights on Bias and Concept Drift |
| | |
| | **Abstract**: |
| | SHAP values have become a cornerstone in explaining machine learning models by providing insights into the importance of individual features. In this talk, we delve into the application of SHAP values in two challenging scenarios: imbalanced binary classification and longitudinal problems with concept drift. First, we explore how data imbalance affects SHAP values, potentially introducing biases in feature importance assessments. Then, we discuss an ongoing project leveraging SHAP values to intuitively analyze types of concept drift in longitudinal datasets, processed in chunks, within imbalanced scenarios. This work highlights the potential of SHAP values to uncover meaningful trends and drift types through innovative visual representations. Additionally, the projects presented are part of collaborative efforts with GEIST researchers, fueled by a three-month PhD research stay in Krakow. |
| | |
| | **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> |
| | <WRAP clear></WRAP> |
| | |
| | ==== 2024-12-19 ==== |
| | <WRAP column 15%> |
| | {{ :aira:betül-bayrak-foto.jpg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Betül Bayrak, PhD Candidate @ Norwegian University of Science and Technology |
| | |
| | **Title**: Post-hoc XAI Methods: Counterfactuals and XCBR Applications |
| | |
| | **Abstract**: |
| | This talk provides insights into the development of explainable AI (XAI) frameworks, with a particular focus on post-hoc explanation methods and their applications. Emphasizing the role of counterfactuals and the utilization of case-based reasoning methodologies in XAI, the presentation explores how tailored explanations enhance model comprehensibility, support actionable decision-making, and foster trust in AI systems. Additionally, it addresses challenges in evaluating instance-based explanations, integrating human-centered design principles, and balancing technical performance with user-centric needs. Through practical use cases and a discussion of key evaluation metrics, this session aims to bridge the gap between technical innovation and user trust in XAI. |
| | |
| | **Biogram**: |
| | Betül Bayrak is a PhD candidate in the Computer Science Department at the Norwegian University of Science and Technology (NTNU), specializing in Explainable AI (XAI). Doctoral research focuses on instance-based explanations and their evaluation, with a particular emphasis on counterfactual explanations, bridging technical advancements with user experience and applying expertise across diverse domains. Betül's research aims to develop frameworks that enhance comprehensibility, trustworthiness, and practical usability in AI systems, contributing to the XAI literature. |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | ==== 2024-12-12 ==== |
| | <WRAP column 15%> |
| | {{ :aira:jan-ignatowicz-foto.jpg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Jan Ignatowicz, PhD Candidate @ Jagiellonian University |
| | |
| | **Title**: Knowledge Graphs for Digitized Manuscripts in Cultural Heritage Applications |
| | |
| | **Abstract**: |
| | The integration of knowledge graphs with digitized manuscripts opens new possibilities for cultural heritage applications by enabling structured and enriched representations of historical data. This presentation explores an ongoing project utilizing pre-trained deep learning models, fine-tuned on a custom dataset specifically curated for this task. These models extract additional information from digitized manuscript images, such as identifying the presence and location of textual elements, decorative patterns, stamps, and other significant features. |
| | By combining these extracted details with pre-existing metadata, comprehensive knowledge graphs are constructed to provide a semantic, interconnected view of the digitized manuscripts. These knowledge graphs aim to support advanced querying, reasoning, and interoperability within cultural heritage collections. The project demonstrates how these enriched digital resources can foster a deeper understanding of historical manuscripts while creating a foundation for their integration into broader cultural and academic frameworks. |
| | |
| | |
| | **Biogram**: |
| | Jan Ignatowicz (jan.ignatowicz@doctoral.uj.edu.pl) is a PhD candidate at the Jagiellonian University in Technical Computer Science since 2023. He received BCs degree from Computer Science at the Faculty of Electrical Engineering, Automatics, Computer Science and Biomedical Engineering of the AGH University of Science and Technology in 2020, and MSc degree from Computer Science at the Faculty of Mathematics and Computer Science of the Jagiellonian University in 2022. |
| | He is mainly interested in machine learning and its applications in fields of Image Understanding and Natural Language Understanding. |
| | In free time he enjoys ballroom dance, traveling and learning foreign languages. |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | ==== 2024-12-05 ==== |
| | <WRAP column 15%> |
| | {{ :aira:ulvi-movsum_zada-foto.jpg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Ulvi Movsum-zada, PhD Candidate @ Jagiellonian University |
| | |
| | **Title**: VisTabNet: Adapting Vision Transformers for Tabular Data |
| | |
| | **Abstract**: |
| | In this talk, I will present VisTabNet, an innovative approach that adapts Vision Transformers (ViTs) to process tabular data, leveraging the power of transfer learning across modalities. VisTabNet reuses pre-trained ViTs by projecting tabular inputs into patch embeddings, effectively using an adaptation network to make this cross-modal transfer possible. Our experiments demonstrate that VisTabNet not only reduces the architectural complexity involved in model design but also significantly decreases training costs while showing strong performance on small tabular datasets. I will discuss its applicability in various real-world scenarios, particularly where traditional deep learning models have struggled to outperform ensemble methods on tabular data. |
| | |
| | **Biogram**: |
| | Ulvi Movsum-zada holds two master’s degrees, one in Management and another in Nanotechnology and Advanced Materials. Currently, he is in the first year of a PhD in Information Technology in UJ. Ulvi works as a Senior Software Engineer at IBM on the AutoAI project. Alongside this, he leads his own NGO, "Chapter of Enlightenment," where he manages research groups in three domains: Artificial Intelligence, Quantum Computing, and Space. |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | ==== 2024-11-28 ==== |
| | <WRAP column 15%> |
| | {{ :aira:michał-wierzchon-foto.jpeg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Michał Wierzchoń, Professor @ Jagiellonian University |
| | |
| | **Title**: Centre for Brain Research – aims, goals, case projects and available datasets |
| | |
| | **Abstract**: |
| | The field of human neuroscience is undergoing a significant transformation. Researchers have recognized that small-scale, single method experiments with limited number of participants and control over research protocols often yield inconclusive results and are susceptible to false positives. To address these challenges, there is a shift towards leveraging large-scale datasets obtained through multi lab studies using multiple neuroscience methods. |
| | |
| | In line with this trend, we have established the Centre for Brain Research, an interdisciplinary unit of the Jagiellonian University dedicated to analyzing human brain data in both health and disease. This presentation will outline the Centre's aims and goals, provide an overview of the neuroscience methods we utilize, as well as the typical and advanced data analysis protocols we employ. |
| | |
| | We will also showcase recent research projects and the corresponding datasets we have collected at the Centre, which include studies on the impact of blue light and sleep deprivation on brain function (A. Domagalik), the effects of psychedelics, neuronal mechanisms underlying perception (M. Bola), as well as consciousness, or even the perception of contemporary classical music (M. Wierzchon). |
| | |
| | Our focus will be on potential follow-up analyses that necessitate interdisciplinary collaboration. |
| | |
| | **Biogram**: |
| | My main research interest is the topic of consciousness viewed in the perspective of cognitive neuroscience. My research projects focus on conscious perception, memory and self-awareness. I use various methods of measuring consciousness, combining the methodology of behavioral, neurobiological, qualitative research and computer simulations. Furthermore, I am interested in the application of the results of the above research in the clinical context (diagnosis and treatment of disorders of consciousness and body representation), including rehabilitation practice, especially concerning the development of perception-enhancing tools and brain-computer interfaces. |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | |
| | ==== 2024-11-21 ==== |
| | <WRAP column 15%> |
| | {{ :aira:agnieszka-lawrynowicz-foto.jpg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Agnieszka Ławrynowicz, Associate Professor @ Poznan University of Technology |
| | |
| | **Title**: How to Build Trustworthy Knowledge Graphs? |
| | |
| | **Abstract**: |
| | In this presentation, I will introduce the topic of knowledge graph extraction and construction, and outline the typical process leading to a knowledge graph ready for application. A knowledge graph is a modern form of knowledge representation as a semantic network with added constraints. In a knowledge graph, the data schema, often based on ontologies, represents only a small part, while the focus is on instances, which can be vast in number. This scale requires appropriate knowledge acquisition methods, utilizing information extraction, data mining, and machine learning, presenting challenges related to verifying correctness, completeness, consistency, and knowledge freshness. The presentation will discuss these challenges and corresponding solutions, illustrated with examples from various projects I have participated in or led, including those in the fields of digital humanities, food, and machine learning experiment modeling. |
| | |
| | **Biogram**: |
| | Agnieszka Ławrynowicz is an Associate Professor at the Faculty of Computer Science and Telecommunications, Poznan University of Technology and head of the Semantics and Knowledge Engineering Laboratory. She is a member of the Scientific Council of the Polish Association for Artificial Intelligence, ECCAI, programe and organising committees of leading international conferences in the field of artificial intelligence and knowledge engineering (e.g. ISWC, K-CAP, EKAW, WWW, ECAI), chair of the Knowledge Engineering track at the conference of the Polish Association for Artificial Intelligence and member of the Editorial Committees of the journals Transactions on Graph Data and Knowledge and Semantic Web. She has led or participated in a number of research projects funded by the European Commission, Norwegian funds, FNP, NCN, NCBiR, and as a member of the TAILOR European network of research laboratories on the topic of trustworthy artificial intelligence based on the integration of reasoning, learning and optimisation. She was a scholarship holder in the Marie-Curie programe of the European Commission for a project on web mining at the University of Ulster, a winner of a grant in a programe financed by the FNP for a project in collaboration with Stanford University, a winner of an award for an outstanding monograph in computer science awarded by the Committee on Informatics of the Polish Academy of Sciences, a ""Scientist of the Future"" award, a promoter of the most innovative engineering thesis in Poland (competition under the auspices of the IEEE) and other awardees pursuing work in the field of artificial intelligence. She is an expert on ethics at the European Commission. |
| | |
| | Research interests: artificial intelligence, knowledge engineering, knowledge graphs, information extraction, Semantic Web |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | ==== 2024-11-14 ==== |
| | <WRAP column 15%> |
| | {{ :aira:dorota-głowacka-foto.jpg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: Dorota Głowacka, Professor @ University of Helsinki |
| | |
| | **Title**: User-centric Design and Evaluation of Exploratory Search and Recommender Systems |
| | |
| | **Abstract**: |
| | In this talk, I will give an overview of my research interests and describe recent projects in information retrieval by members of my research group. Topics I will discuss include the design of exploratory search systems for scientific literature and generative adversarial networks (GANs), as well as aspects of the evaluation of cross-domain recommendations and sequential recommender systems. In general, this research takes a user-centric approach where we prioritise understanding and enhancing how users interact with information systems, and how offline evaluation fails to capture important aspects of real-world use, such as user perception and data leakage. |
| | |
| | **Biogram**: |
| | Dorota Głowacka is an Associate Professor in Artificial Intelligence in the Department of Computer, University of Helsinki. She completed and MSc and PhD in Computer Science at University College London (UCL). Prior to her current appointment, she was an Assistant Professor in Machine Learning in the School of Informatics, University of Edinburgh. Her research interests are in interactive information retrieval, recommender systems and human-AI interaction. |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | |
| | ==== 2024-11-07 ==== |
| | <WRAP column 15%> |
| | {{ :pub:whryniewska_zdj.jpg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: M.Sc. Eng. Weronika Hryniewska-Guzik |
| | |
| | **Title**: NormEnsembleXAI: Unveiling the Strengths and Weaknesses of XAI Ensemble Techniques |
| | |
| | **Abstract**: |
| | During this seminar, I will provide a comparative analysis of explainable artificial intelligence (XAI) ensembling methods. The presentation will highlight three key contributions. First, I will introduce NormEnsembleXAI, a novel ensembling method that combines aggregation functions—specifically, the minimum, maximum, and mean—combined with normalization techniques to improve interpretability. Second, I will discuss the strengths and weaknesses of existing XAI ensemble approaches, offering valuable insights into their practical applications. Lastly, I will showcase a new library designed to facilitate the implementation of XAI ensembling, supporting the development of more transparent and interpretable deep learning models. |
| | |
| | **Biogram**: |
| | Weronika Hryniewska-Guzik has submitted her Ph.D. thesis in Computer Science at the Warsaw University of Technology. Her research focuses on deep learning, exploring aspects such as segmentation, representation learning, and multi-tasking on medical images. With a particular emphasis on human-oriented explanations, she investigates methods like human-in-the-loop explanations and explanations ensembling. She gained experience during laboratory internships at Taiwan Tech and is currently doing a 3-month research visit at Nanyang Technological University in Singapore. |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | |
| | ==== 2024-10-24 ==== |
| | <WRAP column 15%> |
| | {{ :aira:lucjanjanowskipub.jpg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: dr hab. inż. Lucjan Janowski |
| | |
| | **Title**: Quality of Experience - Quality for Telecommunication |
| | |
| | **Abstract**: |
| | Quality of Experience (QoE) is a concept developed in the context of network optimization, where the challenge lies in aligning technical metrics with human perception. This lecture will begin by discussing general strategies for network optimization and their impact on user satisfaction. I will then delve into the complexity of collecting subjective scores to measure QoE. Unlike objective metrics such as latency or throughput, subjective scores reflect personal user experiences, making the design of experiments more intricate. We will examine best practices for conducting these subjective experiments, ensuring consistency and reliability in QoE studies, and explore how they differ from traditional user experience (UX) research. Finally, I will introduce emerging areas like biofeedback and virtual reality, which offer new opportunities for enhancing QoE measurement. Biofeedback can provide real-time insights into a user’s physiological state, offering a deeper understanding of their experience, while virtual reality is increasingly important for QoE assessment due to its immersive nature. These technologies represent the future of QoE research, pushing the boundaries of how we perceive and measure user satisfaction in digital environments. |
| | |
| | **Biogram**: |
| | Lucjan Janowski received his Ph.D. in Telecommunications from the AGH University of Science and Technology, Krakow, Poland, in 2006. In 2007, he was a postdoctoral researcher at the Laboratory for Analysis and Architecture of Systems, Centre National de la Recherche Scientifique, in Paris, France. From 2010 to 2011, he continued his postdoctoral research at the University of Geneva, Switzerland. From 2014 to 2015, he was a postdoctoral researcher at the Telecommunications Research Center Vienna, Austria. He is currently an assistant professor at the Institute of Telecommunications, AGH University of Science and Technology. His research interests include statistical and probabilistic modeling of subjects and subjective ratings used in Quality of Experience (QoE) evaluation. https://scholar.google.com/citations?hl=en&user=ZWyd1YYAAAAJ |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| | |
| | |
| | ==== 2024-10-17 ==== |
| | <WRAP column 15%> |
| | {{ :pub:about_us:lvm.jpg?200| }} |
| | </WRAP> |
| | |
| | <WRAP column 75%> |
| | |
| | **Speaker**: dr Luiz do Valle Miranda |
| | |
| | **Title**: Critical Early-Stage Decisions in Linked Data Projects for Cultural Heritage: Challenges, Choices, and Guidelines |
| | |
| | **Abstract**: |
| | Linked Data (LD) has attracted growing interest in the Gallery, Library, Archive, and Museum (GLAM) sector due to its ability to link the spatial, temporal, and relational contexts of Cultural Heritage (CH) resources—such as books, journals, and drawings. By enhancing metadata interoperability, LD improves accessibility and fosters collaboration across CH institutions. However, several barriers still prevent the wider adoption of LD, and addressing these obstacles is crucial to realizing LD's full potential within the sector, one of them being confusion on critical decision-making in the early stages of CH projects. |
| | |
| | In this presentation, I want to outline some of the key early stage critical points in LD-based CH projects, and the possible alternative decisions for these critical points. By discussing the impact of such a choice, I aim at highlighting the usefulness of more explicit guidelines for such decision-making. Finally, I present how I aim at contributing towards the exposition of such guidelines by means of a survey of LD-based CH projects. The survey will focus on understanding how these critical points were dealth with in psat and currency CH projects. |
| | |
| | **Biogram**: |
| | Luiz do Valle Miranda holds a Ph.D. in Philosophy from Charles University in Prague and has been a Ph.D. candidate in Technical Computer Science at Jagiellonian University since 2023. He earned his BA in Cognitive Science from the John Paul II Catholic University of Lublin in 2018 and his MA in Philosophy from the University of Antwerp in 2020. His current academic research is centered on the development of knowledge-based systems, particularly in the domain of Cultural Heritage. His interest lies in the intersection between machine learning methods and explicitly represented background knowledge towards the creation of intelligent technology that fosters care and human flourishing. |
| | </WRAP> |
| | <WRAP clear></WRAP> |
| |
| ==== 2024-06-27 ==== | ==== 2024-06-27 ==== |
| |
| |
| **Title II**: Context-Aware learning models: CALM-Project by José Palma, Juan Botía, University of Murcia | **Title I**: Context-Aware learning models: CALM-Project by José Palma, Juan Botía, University of Murcia |
| |
| **Abstract**: | **Abstract**: |
| These are the main challenges that CAML project tries to approach. The initial hypothesis of this project is that the development of advanced techniques to detect and characterise concept drift in pandemic scenarios and on different types of data, combining different representation schemes, will not only provide tools to anticipate and react to model degradation, but will also allow the development of techniques to generate medical and preventive knowledge that will help improve the quality of care and, ultimately, better manage situations of extreme stress in the health system. | These are the main challenges that CAML project tries to approach. The initial hypothesis of this project is that the development of advanced techniques to detect and characterise concept drift in pandemic scenarios and on different types of data, combining different representation schemes, will not only provide tools to anticipate and react to model degradation, but will also allow the development of techniques to generate medical and preventive knowledge that will help improve the quality of care and, ultimately, better manage situations of extreme stress in the health system. |
| |
| | |
| | **Title II**: Concept Drift in Imbalanced Problems by Antonio Guillén-Teruel, University of Murcia |
| | |
| | **Abstract**: |
| | We address the challenge of concept drift in imbalanced datasets, which is common in various real-world applications. We build upon the IPIP (Identical Partitions for Imbalance Problems) method, which effectively generates balanced subsets by subsampling the majority class to ensure representation of all minority class instances. This method enhances the performance of ensemble learning models in imbalanced scenarios. |
| | |
| | Additionally, we introduce the UIC (Unbiased Integration Coe) metric, designed to reduce bias caused by the imbalance ratio in datasets. The UIC metric integrates multiple biased measures, inversely weighted by their correlation with the minority class proportion, resulting in a more unbiased evaluation metric. |
| | |
| | Our work extends these methods to address concept drift, particularly in the context of passive learning. Concept drift occurs when the underlying data distribution changes over time, a phenomenon observed in dynamic datasets like those tracking COVID-19 patient outcomes. We propose adapting the IPIP method to handle concept drift by updating the model with new data chunks over time, ensuring that minority class instances are adequately represented in each update. |
| | |
| | Furthermore, we explore the application of the UIC metric to problems involving concept drift in imbalanced data. By adapting UIC for these scenarios, we aim to provide a more reliable measure of model performance over time, despite the evolving data distributions. |
| | |
| | Through extensive experiments on both simulated and real-world datasets, including a COVID-19 patient cohort, we demonstrate the effectiveness of our approach. Our findings suggest that the combination of IPIP and UIC, adapted for concept drift, offers a robust framework for tackling imbalanced data in non-stationary environments. Future work will focus on developing R packages to implement these methods, facilitating their application in various practical settings. |
| |
| **Biograms**: | **Biograms**: |
| |
| **Juan A. Botía**, is Professor in Computer Science and Artificial Intelligence at University of Murcia (UMU), Spain since September 2018. He also holds an Honorary Senior Research Fellow position at the Institute of Neurology (IoN), University College London (UCL), UK since July, 2017. He has a PhD in Computational Science and Artificial Intelligence (AI) from UMU (March, 2002). His expertise combines a deep knowledge about Artificial Intelligence, the experience of 10 years in the development of bioinformatic pipelines and the application of machine learning on the molecular biology domain. During the last 10 years, he has actively participated in 43 scientific papers indexed at Journal Citation Reports. Within that period he has supervised and completed 7 PhD theses. He is currently supervising 4 PhD students (4 at University of Murcia, 1 at University College London). He has experience of more than 15 years teaching machine learning related subjects and six years of teaching data analysis for bioinformatics. | **Juan A. Botía**, is Professor in Computer Science and Artificial Intelligence at University of Murcia (UMU), Spain since September 2018. He also holds an Honorary Senior Research Fellow position at the Institute of Neurology (IoN), University College London (UCL), UK since July, 2017. He has a PhD in Computational Science and Artificial Intelligence (AI) from UMU (March, 2002). His expertise combines a deep knowledge about Artificial Intelligence, the experience of 10 years in the development of bioinformatic pipelines and the application of machine learning on the molecular biology domain. During the last 10 years, he has actively participated in 43 scientific papers indexed at Journal Citation Reports. Within that period he has supervised and completed 7 PhD theses. He is currently supervising 4 PhD students (4 at University of Murcia, 1 at University College London). He has experience of more than 15 years teaching machine learning related subjects and six years of teaching data analysis for bioinformatics. |
| | |
| | **Antonio Guillén-Teruel** 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> | </WRAP> |
