Machine Learning and Algorithms in Experimental Particle Physics (MVSem)

Date: Fridays, 11:15-13:00

First meeting: Friday, 16.10.2026 (Introduction, distribution of topics)

Place: NF 226, EG, Goldbox K2 + K3

Registration: via heiCO

This seminar introduces machine learning techniques as well as classical algorithms, and explores their applications in contemporary particle physics experiments such as ATLAS, ALICE, CMS, and LHCb. Each session will begin with a clear and accessible overview of a specific algorithm, followed by a discussion of how it is used in actual high-energy physics analyses. The seminar is designed to be approachable for students from a variety of backgrounds. Prior experience with machine learning is helpful but not required.

Topics include:

Particle Identification with Boosted Decision Trees and Neural Networks

Bayesian Parameter Estimation

Particle Tracking with the Kalman Filter

Detector simulation with GEANT

Generative Models for Detector Simulation

Graph Neural Networks for Particle Tracking and Reconstruction

Anomaly Detection

Fast Machine Learning for Triggering and Data Acquisition

Uncertainty Quantification in ML Predictions

Unfolding in Particle Physics: From Traditional Methods to Machine Learning

Symbolic regression

Jet tagging with Transformers