Deep Learning

by Karolos Potamianos (University of Warwick (GB))

Monday 17 Aug 2026, 09:00 → Tuesday 18 Aug 2026, 16:00 Europe/Zurich

593/R-010 - Salle 11 (CERN)

Abstract

This class focuses on Deep Learning algorithms, covering a brief intro on the topic, a discussion on training strategies (gradient descent, and optimisation and regularisation strategies). Starting from Multi-Layer Perceptrons, it presents the underlying concepts before looking at more complex architectures such as convolutional, recurrent, and graph neural networks. Finally, applications for each of thee are presented.

Lecturer Bio 

Karolos Potamianos is Associate Professor at the University of Warwick (UK). He earned his PhD from Purdue University, focusing on searches for the Higgs boson in association a W or Z boson at CDF, using deep learning, when the term started to become popular, but before modern frameworks existed. Since 2012, working successively at LBNL (US), DESY (DE), Oxford (UK and Warwick (UK), he’s been working on the ATLAS experiment, both on detector integration, commissioning and operation, and on LHC data analysis using deep learning.