Alina Bialkowski

Dr Alina Bialkowski is a Senior Lecturer at the School of Electrical Engineering and Computer Science, part of the Faculty of Engineering, Architecture and Information Technology at The University of Queensland. Her research focuses on interpretable machine learning and computer vision to enhance AI transparency and solve real-world challenges. Prior to joining UQ in late 2017, she held postdoctoral positions at University College London (2015–2017), where she studied human perception in driving, and Disney Research Pittsburgh (2014), analyzing team sports using spatiotemporal data.

Dr Bialkowski earned her PhD and Bachelor of Engineering (Electrical Engineering) from Queensland University of Technology, Australia. Her doctoral research centered on group behavior analysis from visual and spatiotemporal data, with applications in sports analytics and intelligent surveillance systems.

Her research interests span medical imaging (especially electromagnetic imaging of strokes), human attention modeling in driving, intelligent transport systems, surveillance systems, and sports analytics. She emphasizes explainable AI to bridge the gap between technical systems and human understanding, employing methods like feature visualization and attribution. Her work also explores sensors for non-invasive imaging and machine learning frameworks to ensure ethical AI.

Dr Bialkowski has received significant recognition, including the Best Paper Prize at the 2017 IEEE Winter Conference on Applications of Computer Vision (WACV). Her research has led to 6 international patents with collaborators such as Disney Research, Toyota Motor Europe, and The University of Queensland, focusing on electromagnetic imaging and AI-driven solutions.

She actively contributes to interdisciplinary projects like The Lanyard Project (traffic sensor fusion) and co-authored a SmartSat CRC-funded research report on machine learning for satellites. Dr Bialkowski is available for supervision and advocates for AI systems that integrate human-centric principles into their design and evaluation.