Research |
Widely Applied Math
The MIGHTE Lab develops advanced mathematical models and machine learning algorithms to address complex problems in optimization, data processing, and the prediction of dynamic system behaviors. Our approach integrates mathematical analysis, statistics, and computational techniques to create efficient and scalable solutions in areas such as healthcare and data science. By leveraging methods like supervised and unsupervised learning and deep learning, we aim to uncover meaningful patterns from large datasets and improve data-driven decision-making.Additionally, the lab focuses on implementing mathematical models to interpret and enhance the performance of machine learning algorithms, ensuring their stability, robustness, and computational efficiency. We engage in research on convex optimization, probability theory, and stochastic processes to develop innovative methods for neural network training, dimensionality reduction, and uncertainty modeling. Through interdisciplinary collaboration, we aim to transform theoretical knowledge into practical applications that propel technological and scientific advancements.
Related publications
|
Parameter Inference of Black Hole Images using Deep Learning in Visibility Space
Franc O, Pavlos Protopapas, Dominic W. Pesce, Angelo Ricarte, Sheperd S. Doeleman, Cecilia Garraffo, Lindy Blackburn, Mauricio Santillana arXiv:2504.21840 [astro-ph.GA] https://doi.org/10.48550/arXiv.2504.21840 |
|
Unveiling individual and collective temporal patterns in the tanker shipping network
Kevin Teo, Naomi Arnold, Andrew Hone, Michael Coulon, Martin Ireland, Mauricio Santillana, István Z. Kiss. arXiv:2502.19957v1 [physics.soc-ph] https://doi.org/10.48550/arXiv.2502.19957 |
