UCLA

Jets comprise a rich class of emergent phenomena stemming the underlying theory of the strong nuclear force, Quantum Chromodynamics. As jets are produced in copious quantities in hadron colliders, understanding their internal structure and evolution is of the utmost importance for modern particle physics. In this thesis, we study various aspects of a special class of jets---that is, jets containing heavy quarks, such as charm, bottom and top---which can all be understood from a statistical point of view. In the first part, we consider situations in which the observation of back-to-back heavy-quark dijet pairs shed light on key physics governing the final and initial states of high-energy particle collisions---from the modification of dijet mass spectra by the quark-gluon plasma created in the collisions of heavy ions to the probing of the Sivers spin asymmetry in deep inelastic scattering. In the second part, we analyze the internal landscapes of jets initiated by heavy quarks and demonstrate how the so-called ``dead-cone'' effect manifests in the cumulants of jet substructure distributions. In the third and final part, we adapt concepts from the machine learning community to tag top jets from a background of jets initiated by light quarks and gluons as well develop a novel data type that is particularly well-suited to exposing the characteristic angular structure of top decay products.