UPDATE: From 27 May 2023 onwards, the dissertation draft will be removed from the repository as it is currently undergoing assessment by the Examiner at the University of Oxford. The final submitted version of the dissertation will be uploaded once the assessment process has been completed.
This repository is for an ONGOING master dissertation titled: "Non-parametric Inference of Gravitational Potential of a Galaxy from a Kinematical Snapshot". The expected finishing date would be on 29 May 2023, but so far, note that the project is still INCOMPLETE!!!
The ongoing dissertation draft contains an incomplete but self-consistent methodology on a non-parametric framework to infer the gravitational potential of a galaxy from a kinematical snapshot, where Gaussian process regression is used as the key statistical method.
The Scripts directory contains an ongoing code for an application of the framework to a toy model galaxy of isothermal disk.
Users will need to install Tensorflow and GPflow Python packages in order to run the code in this repository.