From 21 to 25 September 2026, Białystok University of Technology hosted the 9th German-Polish Conference on Optimization – Methods and Applications (GPCO 2026). Lublin University of Technology was represented by Karol Łazaruk and Piotr Tokarski of the Department of Computer Science, who presented the results of their research.
The conference is part of a series of Polish-German scientific meetings held since 1999, alternating between Poland and Germany. This year’s programme covered optimization methods and their applications, with particular attention to optimization in machine learning and the use of fractional calculus. The event also included workshops for doctoral students led by invited lecturers from Poland and Germany.
Karol Łazaruk, MSc Eng., presented “Generalized Choquet Integrals for Enhancing Classical Machine Learning via Non-Additive Feature Modeling in Eye-Tracking-Based UX Behavioral Classification.” His work applied generalized Choquet integrals to aggregate the outputs of machine learning classifiers in recognizing user behavior from eye-tracking data.
Piotr Tokarski delivered “Eyes on the Cube: Optimized Feature Selection for Gaze-Based Classification of Pre-Inspection and Solving Phases in Speedcubing.” His research focused on identifying the inspection and solving phases of a Rubik’s Cube attempt from eye movements, with a key contribution being the selection of a minimal feature set that still distinguishes the two phases reliably.
Both talks combined eye-tracking analysis with machine learning and optimization methods, showing how these approaches can support research on human behavior.
Taking part in GPCO 2026 gave our researchers the opportunity to present their findings to an international audience, exchange experiences with colleagues from Poland and Germany, and build scientific contacts that may lead to future collaboration.
Conference website: https://gpco2026.pb.edu.pl/
