Using Machine Learning for Model-Independent New Physics Discovery at the Large Hadron Collider

Despite compelling theoretical and observational evidence for new physics phenomena, direct searches at the Large Hadron Collider (LHC) have not yet discovered any physics Beyond the Standard Model (BSM). While the sensitivity and reach of direct searches for BSM models at the LHC has been steadily increasing, there is a growing need for search methods capable of discovery in unexpected scenarios. A promising avenue of recent research to this end has used machine learning to construct new model-independent search methods. In this thesis, I present an analysis of several such methods ranging over a variety of model dependencies, finding that a measured combination of approaches has the potential to broadly increase the reach of the new physics search program at the LHC.