LHC Physics as Fun Data Science
- Colloquium

Machine Learning for Particle Theory
Modern machine learning is having significant impact on essentially all aspects of LHC physics. The simple reason is that LHC physics uniquely combines vast and highly complex data sets with precise first-principles predictions. I will introduce some ML-related aspects of LHC theory and show how we can benefit from new concepts and methods. This includes precision simulations including uncertainty estimates, inverted simulations and unfolding in a mathematically consistent manner, anomaly searches, and symbolic regression, to close the theory circle and return to formulas.
![Band structure of 2D semimetal based on HgTe quantum well. Experimental points are obtained from the analysis of the cyclotron resonance in the quasi-classical approximation. Solid lines are predictions of the kp theory with no free parameters. Splitting of the conduction (e1,2) and valence (h1) band is due to the quantum confinement. [J. Gospodaric, AP, et al., PRB 104, 115307].](/storages/physik/_processed_/b/5/csm_Kolloquium_Pimenov_0fa7761647.png)





