The goal of this thesis was to explore the space of gravitational wave analysis using deep learning approaches. At the time of writing this thesis, a machine learning challenge was being held by the Albert-Einstein-Institut Hannover and the Friedrich-Schiller Universität Jena with the goal of capturing the state of the art and finding new useful approaches. This thesis makes heavy use of the material provided by this challenge.
MLGWSC: Contains codes to reproduce the basic example given by the machine learning challenge mentioned above. It also contains it as a submodule.MatchedFilter_ext&matchedFilter: Contains implementations of matched filtering.scripts: Contains all the scripts used to define and train the network, generate data and evaluate the results..tex: Contains the source for the actual written thesis.
This thesis was graded with 6.0