School of Particles and Accelerators

Journal Club: Experiments & Phenomenology / Hybrid Format



Date:

Monday, 5 May 2025/ دوشنبه، 15 اردیبهشت 1404

Time:

11:00 AM - 12:00 PM (Tehran time)

Speaker:

Dr. Meisam Ghasemi


Affiliation:

School of Particles and Accelerators, IPM

Title:

Applications of Classical and Quantum Neural Networks in HEP Analyses


Title in Persian:

کاربردهای شبکه عصبی کلاسیک و کوانتومی در تحلیل های فیزیک انرژی بالا



Abstract:

This presentation explores the application of classical and quantum machine learning techniques in particle physics, highlighting their role in tasks such as jet classification, Higgs boson and SUSY identification, and fully hadronic event generation. It begins with an overview of the historical development of artificial intelligence and foundational concepts in deep learning, followed by practical implementations of deep neural networks (DNNs) and convolutional neural networks (CNNs) in high-energy physics analyses, including supersymmetry searches and jet substructure studies. The presentation then introduces quantum machine learning (QML), covering models such as quantum support vector machines (QSVMs) and variational quantum classifiers (VQCs), along with their advantages and challenges in Higgs boson identification. Finally, the use of generative AI is presented as a solution for data generation in both Standard Model and beyond Standard Model scenarios, enabling the production of synthetic datasets significantly faster than traditional Monte Carlo simulations involving MadGraph, Pythia, and Geant4.

Based on: https://link.springer.com/article/10.1140/epjc/s10052-023-11543-6 ,

https://cs231n.stanford.edu/reports/2016/pdfs/300_Report.pdf ,

https://arxiv.org/abs/2104.07692,

https://arxiv.org/abs/1701.05927


Indico Link: https://indico.hep.ipm.ir/e/m.ghasemi


Meeting Place: Seminar Room, School of Particles and Accelerators, IPM

Link to Join Virtually: https://www.skyroom.online/ch/ipm-particles/journal-club