A Machine Learning Technique for Dynamic Aperture Computation
Résumé
Currently, dynamic aperture calculations of high-energy hadron colliders are performed through computer simulations, which are both a resource-heavy and time-costly processes. The aim of this study is to use a reservoir computing machine learning model in order to achieve a faster extrapolation of dynamic aperture values. A recurrent echo-state network (ESN) architecture is used as a basis for this work. Recurrent networks are better fitted to extrapolation tasks while the reservoir echo-state structure is computationally effective. Model training and validation is conducted on a set of "seeds" corresponding to the simulation results of different machine configurations. Adjustments in the model architecture, manual metric and data selection, hyper-parameters tuning and the introduction of new parameters enabled the model to reliably achieve good performance on examining testing sets.