Matias Jari Johannes Uusinoka
My research centers on developing machine learning and statistical physics frameworks for analyzing the dynamics of complex systems—more specifically ice deformation fields. I currently work on methods for learning dynamics from noisy and discontinuous radar imagery and developing graph learning-based approaches for large-scale discrete element models. I also explore concepts from statistical physics present in ice dynamics (multifractality, renormalization group, self-organized criticality) with the goal of linking engineering-scale ice mechanics to geophysical scale statistical behavior and uncovering scale-dependent structures and critical behavior. I am interested other related research topics across diverse application domains.
Full researcher profile
https://research.aalto.fi/...
Email
matias.uusinoka@aalto.fi
Areas of expertise
Statistical physics, Complex systems, Machine learning, Deep learning, Ice dynamics
Publications
Deep Learning-Based Optical Flow in Fine-Scale Deformation Mapping of Sea Ice Dynamics
Matias Uusinoka, Jari Haapala, Arttu Polojarvi
2025
Geophysical Research Letters
Deep learning for automated surveillance of sea ice dynamics
Matias Uusinoka, Arttu Polojärvi, Jari Haapala
2025
Proceedings of the 28th International Conference on Port and Ocean Engineering under Arctic Conditions
Threshold Domain Sizes for Multifractality in Sea Ice Deformation
Matias Uusinoka, Antoine Savard, Jan Åström, Jari Haapala, Arttu Polojärvi
2025
Geophysical Research Letters