Our mission is to advance the frontiers of machine learning through innovative algorithms and impactful real-world applications in challenging domains. We operate at the intersection of theoretical innovation and practical solutions, developing novel approaches for complex data while ensuring AI systems remain safe, interpretable, trustworthy, and beneficial to humanity. Our interdisciplinary approach extends to applications in healthcare and bioinformatics, where we develop computational tools that enhance diagnostic capabilities and drug response prediction through analysis of biomedical and chemical data.
The key research areas of our lab include:
- Advanced Classification Techniques: class imbalance, multi-label classification, online learning from evolving data streams, feature extraction from unstructured data, and deep learning models for natural language processing.
- Explainable and Trustworthy AI: counterfactual explanations, prototype networks, visualization techniques, rule-based systems, rigorous evaluation frameworks, AI safety, and fairness in machine learning.
- AI Applications in medicine, biology and chemistry: multi-omic data analysis, drug response prediction, protein and nucleic acid structure validation, ligand prediction, predictive and generative models for chemistry.
Team members:
- Jerzy Stefanowski (team leader)
- Dariusz Brzeziński
- Krzysztof Dembczyński (on a long-term leave)
- Wojciech Kotłowski
- Mateusz Lango (on a long-term leave)
- Maciej Piernik
- Robert Susmaga
- Izabela Szczęch
- Marek Wydmuch
PhD Students:
- Marco Grillo
- Jacek Karolczak
- Adam Krawczyk
- Anna Przybyłowska
- Witold Taisner