Colliding Worlds: Modern Computational Methods for Scattering Amplitude Calculations and Responding to Crisis Situations

Precision theoretical predictions for high multiplicity scattering rely on the evaluation of increasingly complicated scattering amplitudes which come with an extremely high CPU cost. For state-of-the-art processes this can cause technical bottlenecks in the production of fully differential distributions. In this thesis we explore the possibility of using neural networks to approximate multi-jet scattering amplitudes and provide efficient inputs for Monte Carlo integration. We begin by focussing on QCD corrections to jets up to one-loop. We demonstrate reliable interpolation when a series of networks are trained on amplitudes that have been divided into sectors defined by their infrared singularity structure. Complete simulations for one-loop distributions show speed improvements of at least an order of magnitude over standard approaches. We extend our analysis to the case of loop-induced diphoton production through gluon fusion and develop a realistic simulation method that can be applied to hadron collider observables. Specifically, we present a detailed study for and scattering problems which are extremely relevant for future phenomenological studies and find excellent agreement with amplitudes generated using traditional methods. In order to provide a useable technology, we present an interface with the ~Monte Carlo event generator. The techniques underlying our machine learning methodology and Monte Carlo event generator simulations are widely applicable in other domains as well. In this thesis we will also discuss the use of machine learning to aid in rapid response to crises situations, and the parallels between multi-particle event generators and multi-agent simulations for modelling the spread of epidemics. In this latter case, we develop a new agent-based model with highly granular resolution and discuss its applications to modelling the spread of COVID-19 in England, and in refugee and internally displaced person settlements to aid data driven decision making.