首页|Studies from Jozef Stefan Institute in the Area of Machine Learning Described (S ystematic evaluation of generative machine learning capability to simulate distr ibutions of observables at the large hadron collider)

Studies from Jozef Stefan Institute in the Area of Machine Learning Described (S ystematic evaluation of generative machine learning capability to simulate distr ibutions of observables at the large hadron collider)

扫码查看
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – Current study results on artificial in telligence have been published. According to news reporting out of the Jozef Ste fan Institute by NewsRx editors, research stated, “Monte Carlo simulations are a crucial component when analysing the Standard Model and New physics processes a t the Large Hadron Collider.” Financial supporters for this research include Javna Agencija Za Raziskovalno De javnost Rs. The news editors obtained a quote from the research from Jozef Stefan Institute: “This paper aims to explore the performance of generative models for complement ing the statistics of classical Monte Carlo simulations in the final stage of da ta analysis by generating additional synthetic data that follows the same kinema tic distributions for a limited set of analysis-specific observables to a high p recision. Several deep generative models are adapted for this task and their per formance is systematically evaluated using a well-known benchmark sample contain ing the Higgs boson production beyond the Standard Model and the corresponding i rreducible background. The paper evaluates the autoregressive models and normali zing flows and the applicability of these models using different model configura tions is investigated. The best performing model is chosen for a further evaluat ion using a set of statistical procedures and a simplified physics analysis.”

Jozef Stefan InstituteCyborgsEmergin g TechnologiesMachine Learning

2024

Robotics & Machine Learning Daily News

Robotics & Machine Learning Daily News

ISSN:
年,卷(期):2024.(Oct.2)