Andrius Tamosiunas
Testing modified gravity with X-Ray and weak lensing data
Andrius' PHD is focussed on experimental testing of models of modified gravity on galaxy cluster scales. Large quantities of data on X-Ray surface brightness and the strength of gravitational lensing effects from galaxy clusters is analysed to test the accuracy of these models.
It also contains a study of the effectiveness of different machine learning techniques designed to emulate cosmological simulations: neural networks, gradient boosting algorithms and GAN algorithms.
"As part of my DISCnet experience I had an amazing opportunity to do a summer project in the African Institute for Mathematical Sciences (AIMS) in South Africa. There I had a chance to apply my experience gained during the DISCnet training towards developing data science algorithms for a new laser scanner for diagnosing malaria."
"In addition, I worked on developing deep learning techniques simulating a synthetic population for the city of Cape Town. The project resulted in an ongoing collaboration with the researchers at AIMS and the techniques I learnt built a foundation for my current research in cosmology where I use similar techniques to study the large scale structure of the universe."