معرفی
Dr. Sahar Qaadan is a researcher in the Theory of Machine Learning group at the Institute of Neuroinformatics (INI), Faculty of Computer Science, Ruhr University Bochum, Germany. She has been actively working in supervised machine learning and optimization since 2019, following her PhD completion in Tobias Glasmachers' research group at the same institution.
Her educational background includes:
- MSc in Automation and Robotics from Technical University Dortmund, Germany (2011-2014)
- BSc in Mechatronics Engineering from University of Jordan, Amman, Jordan (2003-2008)
Dr. Qaadan's research focuses on developing efficient algorithms for supervised machine learning, particularly optimization techniques for Support Vector Machine (SVM) training. Her work addresses computational challenges in large-scale learning through innovative approaches to budget maintenance, stochastic gradient methods, and convergence acceleration strategies.
Analysis of her 2018-2019 publications reveals a consistent emphasis on enhancing SVM training efficiency. Key contributions include multi-merge budget maintenance techniques, precomputed search methods like Golden Section Search, and specialized adaptations for both stochastic gradient descent and coordinate ascent frameworks. These works collectively target resource-constrained environments while maintaining model accuracy.
No scientific awards were mentioned in the provided documentation.
Information regarding student advising or research grants is not documented in the available materials.
She operates within Tobias Glasmachers' Theory of Machine Learning group at INI, with prior industrial collaboration experience including BMW's Autonomous Systems ADAF project and thermal management control systems for data centers using neural networks and PLC technology.
