Research on Fairness in Machine Learning Models
MISANUAbstract
In recent years, fairness in machine learning models has become a widely discussed research topic due to their increasing application in socially significant and sensitive domains. This lecture presents research conducted within the ONR project, focusing on methodologies for measuring, analyzing, and improving fairness in machine learning models. It addresses key challenges in defining and assessing fairness, as well as developed approaches that incorporate fairness throughout all stages of model development, including data preprocessing, model adaptation, and post-processing adjustments. The goal is to highlight the importance of balancing accuracy, fairness, and transparency in modern machine learning algorithms.
On February 11, 2025, at 14:15, I will deliver an invited lecture at the IEEE Computer Chapter Co-16 Seminar, a long-running collaboration between the Mathematical Institute SANU, the Faculty of Organizational Sciences, and the IEEE C-16 Chapter.
The lecture, titled Research on Fairness in Machine Learning Models, will focus on the importance of fairness in modern machine learning models, covering challenges, methodologies, and research findings from the ONR project. Key topics include measuring, analyzing, and improving fairness in machine learning, with a discussion on methodologies that ensure fairness at all stages of development—from data preprocessing to model adjustments and decision post-processing.
Sandro Radovanović
Assistant Professor at University of Belgrade
My research interests include machine learning, development and design of decision support systems, decision theory, and fairness and justice concepts in algorithmic decision making.