Machine learning for scientific visualization: ensemble data analysis
| PhD ceremony: | H. Gadirov, MSc |
| When: | October 14, 2025 |
| Start: | 16:15 |
| Supervisor: | prof. dr. J.B.T.M. (Jos) Roerdink |
| Co-supervisor: | S.D. Frey, Dr |
| Where: | Academy building RUG / Student Information & Administration |
| Faculty: | Science and Engineering |
Scientific simulations and experimental measurements produce vast amounts of spatio-temporal data, but extracting meaningful insights from such data remains a challenge due to its high dimensionality, complex structures, and missing information. Traditional analysis techniques often struggle with these issues, motivating the need for more robust, data-driven approaches.
In his dissertation, Hamid Gadirov explores deep learning methodologies to enhance the analysis and visualization of spatio-temporal scientific ensembles, focusing on dimensionality reduction, flow estimation, and temporal interpolation. First, Gadirov addresses the challenge of high-dimensional data representation by investigating autoencoder-based dimensionality reduction for scientific ensembles. He evaluates the stability of projection metrics under partial labeling and introduces a Pareto-efficient selection strategy to identify optimal autoencoder variants, ensuring expressive and reliable low-dimensional embeddings.
Next, Gadirov presents FLINT, a deep learning model designed for high-quality flow estimation and temporal interpolation in both flow-supervised and flow-unsupervised settings. FLINT reconstructs missing velocity fields and generates high-fidelity temporal interpolants for scalar fields across 2D+time and 3D+time ensembles, without requiring domain-specific assumptions or extensive fine-tuning. To further improve adaptability and generalization, Gadirov introduces HyperFLINT, a novel hypernetwork-based approach that dynamically conditions on simulation parameters to estimate flow fields and interpolate scalar data. This parameter-aware adaptation enables more accurate reconstructions across diverse scientific domains, even in cases of sparse or incomplete data. By addressing key challenges in scientific data analysis, this dissertation advances deep learning techniques for scientific visualization, providing scalable, adaptable, and high-quality solutions for interpreting complex spatio-temporal ensembles.