UC Santa Barbara

As experimental efforts to uncover the nature of physics beyond the Standard Model continue to push the boundaries of energy and sensitivity, theoretical predictions of well-motivated physics models will need to commensurately increase in precision so that we may reliably distinguish signal from background. A prime example may be found in the stochastic gravitational wave background that will soon be within reach of observatories and which could contain imprints from a variety of ultraviolet phenomena, including first order cosmological phase transitions and topological defects. We begin this thesis by discussing scenarios in which the gravitational wave spectrum due to a phase transition can be substantially altered by particle reflection off of relativistic bubble walls, an effect which has been largely ignored in the literature thus far. We then move on to discussing a particular class of parity-based solutions to the strong CP problem which also features a potential gravitational wave signal, this time due to domain wall topological defects. In addition, these models provide testable predictions for near-future colliders and tabletop experiments. Finally, we point out an exciting new computational avenue for discriminating between signal and background in particle collider data using machine learning coupled with a physically motivated metric on the space of collider events.