Aiden Durrant

Aiden Durrant is a Lecturer in the School of Natural and Computing Sciences at the University of Aberdeen, actively contributing to research and teaching in computing science. He is affiliated with the South China Normal University Joint Institute and is accepting PhD students in Computing Science. His research spans machine learning, computer vision, and AI, with a focus on self-supervised and geometric representation learning.

• BSc (Hons) and MPhil in Computer Science, University of Lincoln
• PhD in Computing Science, University of Aberdeen

Aiden Durrant’s research centers on self-supervised learning and the geometric structuring of learned representations. He investigates how to learn high-quality image representations without human-annotated labels, focusing on preserving semantic and hierarchical knowledge through hyperbolic and hyperspherical geometries. His work extends to multi-modal learning and has practical applications in environmental monitoring, agriculture, nuclear reactor safety, and climate change. He has developed novel architectures such as HMSN (Hyperbolic Self-Supervised Learning) and S-JEA (Stacked Joint Embedding Architectures) to improve representation quality.

His recent publications (2023–2025) demonstrate a strong trend in foundation models, self-supervised learning, and deep learning for environmental and industrial applications. Key themes include crevasse mapping in glaciology, segment anything models for drone imagery, and hyperbolic clustering for semantic hierarchies. His earlier work (2019–2022) focused on nuclear reactor anomaly detection using deep learning and time-series analysis, contributing to the CORTEX Horizon 2020 project.

Scientific Awards:

• PhD Poster Award, FISA - 9th European Commission conference on Euratom research and training in safety of reactor systems (2019)

Durrant has secured competitive funding, including a 2024 SUSTAIN CDT Studentship (PI) and an Aberdeen Internal Pump-Priming grant (Co-PI), supporting research in AI for glacier mapping and sustainable crop breeding. He supervises PhD students and has co-led teaching modules in Machine Learning, Data Mining, and Research Methods at both undergraduate and postgraduate levels. He also contributes nationally as Deputy Director of the SICSA Graduate Academy, supporting research training across all 14 Scottish computer science departments.

He is actively involved in interdisciplinary collaborations, particularly in agri-food data sharing, environmental sustainability, and healthcare AI, often working with cross-institutional teams. His external memberships include BMVA and leadership in organizing the BMVA 2025 Summer School.