New Ideas for LHC Theory using Machine Learning
by
Modern machine learning is having significant impact on essentially all aspects of LHC physics. The reason is that LHC physics uniquely combines vast and highly complex data sets with precise first-principles predictions. I will introduce a range of applications of machine learning to LHC theory and show how we can benefit from these new concepts and methods. Flow-based invertible networks can supplement and enhance precision simulations, including uncertainty estimates and a high level of control through a new, GAN-inspired architecture. Such networks can also unfold detector simulations or the QCD parton shower in a mathematically consistent manner, eventually providing event-wise likelihoods for the matrix element methods. Finally, I will discuss the exciting new avenue of using symbolic regression to learn optimal observables in Higgs physics.
Tilman Plehn is a theoretical physicists at Heidelberg University. His research focus is on searches for new physics at the LHC, including numerical aspects like global analyses and simulations. Together with his group he has been working on applications of modern machine learning to LHC physics since 2017.
M. Girone, M. Elsing, L. Moneta, M. Pierini
EP/IT Data Science Seminar
- Zoom Meeting ID
- 98545267593
- Description
- EP/IT Data Science seminar
- Host
- Lorenzo Moneta
- Alternative hosts
- Thomas Nik Bazl Fard, EP Seminars and Colloquia, Maurizio Pierini, Caroline Cazenoves, Maria Girone, Markus Elsing, Pascal Pignereau
- Passcode
- 97200142
- Useful links
- Join via phone
- Zoom URL
-