Reconstruction of Linear Optics Observables Using Supervised Learning

In the LHC, most of the optical functions can be obtained from turn-by-turn beam centroid data. However, the measurement of such observables as β^{*} and the dispersion function require special dedicated techniques and additional operational time. In this work, we propose an alternative approach to estimate these observables using supervised machine learning, in case the dedicated measurements are not available but turn-by-turn data are. The performance of developed estimators is demonstrated on LHC simulations. Comparison to traditional techniques for the computation of beta-function will be also provided.