Simulation, Data Analysis and Machine Learning in Particle Physics (1/1)

by Dr Artemiy Belousov

Monday 1 Jun 2026, 10:00 → Tuesday 2 Jun 2026, 17:00 Europe/Zurich

N260/3.13

Overview

This training series introduces modern workflows used in particle and nuclear physics, combining Monte Carlo simulations, detector data analysis, and machine learning techniques. The workshops are based on analysis pipelines used in the CBM experiment and reflect real research workflows for studying heavy-ion collisions.
The workshops are designed as a consecutive training track.

Workshop 1 introduces simulation and data analysis workflows, which provide the foundation for Workshop 2, where machine learning methods are applied to scientific datasets.

Location: Takes place on Campus Riedberg of Goethe University Frankfurt in Biozentrum room N260/313. Detailed travel instructions will be provided to participants in advance.

Eligibility: Exclusive to members of academic institutions. The organizers reserve the right to revoke non-eligible registrations!

Registration Fee:  None (exclusive to academic institution members)

Organization: Organized by the NHR-SW.

Daily Timings:  10:00–17:00

Language:  English

Workshop Structure
• Workshop 1: Monte Carlo Simulation and Data Analysis with ROOT
• Workshop 2: Introduction to Neural Networks for Scientific Data Analysis

Workshop Dependency
Workshop 1 provides the necessary understanding of simulation data, detector reconstruction, and physics observables. The datasets and analysis concepts introduced in Workshop 1 will be used as input for machine learning applications in Workshop 2.
Participation in Workshop 1 is strongly recommended before attending Workshop 2.

Workshop 1

Monte Carlo Simulation and Data Analysis with ROOT

Description
This workshop introduces simulation and analysis workflows used in particle and nuclear physics, with examples inspired by the CBM experiment. Participants will learn how Monte Carlo simulations generate collision events using transport models such as PHSD and how reconstruction algorithms, including the Kalman Filter Particle Finder (KFPF), are used to extract physics observables.
The course focuses on practical data processing and analysis using the CERN ROOT framework, including visualization and statistical analysis of simulated and reconstructed event data.

Learning Goals
Participants will learn how to:
• Understand Monte Carlo simulation workflows
• Work with reconstructed detector data
• Use ROOT for data visualization and analysis
• Extract physics observables from collision events

Workshop 2

Introduction to Neural Networks for Scientific Data Analysis

Description
This workshop introduces the fundamental concepts of neural networks with a focus on convolutional neural networks (CNNs). Examples are based on modern data analysis workflows from the CBM experiment, where CNNs are used to classify collision events and identify signatures of quark–gluon plasma formation.
Participants will learn how machine learning models can be trained using simulated datasets generated with transport models such as PHSD and how neural networks can be applied to realistic detector data.

Learning Goals
Participants will learn how to:
• Understand neural network architectures
• Prepare scientific datasets for machine learning
• Train and validate CNN models
• Apply ML methods to detector and simulation data

Two-Day Training Program
Day 1 — Simulation and ROOT Analysis
Morning
• Introduction to heavy-ion collision simulations
• PHSD transport model overview
• Monte Carlo event generation
• Detector reconstruction and KFPF workflow
Afternoon
• Introduction to CERN ROOT
• Event data structure and storage
• Data visualization and statistical analysis
• Hands-on ROOT exercises

Day 2 — Neural Networks and Event Classification
Morning
• Introduction to machine learning in particle physics
• Fundamentals of neural networks
• Convolutional Neural Networks (CNNs)
• Dataset preparation and feature engineering
Afternoon
• CNN training and validation
• Event classification examples from CBM
• Performance evaluation and interpretation
• Hands-on ML exercises