About me
Hello! I'm Ignacy, currently studying Computer Science and Artificial Intelligence at Poznan University of Technology in Poland. My passion lies in the intersection of Machine Learning and transparency, safety, and trustworthiness of AI systems. Currently, I focus a lot on the area of counterfactual explanations.
As a research assistant at the ML Lab in Poznan, my work revolves around enhancing the interpretability of black-box models through robust counterfactual explanations. Currently, I'm deeply involved in developing a statistical approach to generate explanations resilient to model changes, with the aim of making them more actionable for end-users.
My journey in research has been enriched by experiences such as my internship at AutonLab (Robotics Institute, Carnegie Mellon University), where I worked on formal verifications for Bayesian Networks, ensuring critical design specifications were met. Additionally, I was (and am) a part of the R&D team at the Poznan Supercomputing and Networking Center (Polish Academy of Sciences), which provided me with opportunities to work on projects ranging from understanding black-box model behavior to implementing anomaly detection solutions in industrial automotive settings as well as in HPC clusters.
Looking ahead, I plan to further my academic pursuits, whether it's pursuing a Master's or Ph.D. in a research lab dedicated to ML explainability and trustworthiness or exploring other captivating fields that will spark my curiosity.
Research
-
- Ignacy Stępka, Mateusz Lango, Jerzy Stefanowski - "Multi-criteria approach for selecting an explanation from the set of counterfactuals produced by an ensemble of explainers" - International Journal of Applied Mathematics and Computer Science (AMCS) - Mar 20, 2024
-
- Ignacy Stępka, Mateusz Lango, Jerzy Stefanowski - "On usefulness of dominance relation for selecting counterfactuals from the ensemble of explainers" - 4th Polish Conference on Artificial Intelligence, PP-RAI 2023 - Sep 18, 2023
-
- Ignacy Stępka, Nicholas Gisolfi, Artur Dubrawski - "Formal Verification of Bayesian Network Classifiers using Decision Diagrams and Propositional Logic [preprint]" - Carnegie Mellon University, Robotics Institute Summer Scholar Program - Aug 2, 2023