Skip to Content

Team

Summer Fellowships

Procedures

Mailing lists

Last modified by Ricardo Julio Rodríguez Fernández on 2026/06/14 08:33
From version 1.4
edited by Ricardo Julio Rodríguez Fernández
on 2026/06/14 07:45
Change comment: Update document after refactoring.
To version 2.1
edited by Ricardo Julio Rodríguez Fernández
on 2026/06/14 07:48
Change comment: There is no comment for this version

Summary

Details

Page properties
Content
... ... @@ -1,7 +1,5 @@
1 -=== Communicating IGFAE Physics to new audiences ===
2 -==== Supervisor: Manuel Rey Pan
1 +=== Communicating Generative Machine Learning Models for Neutron Tomography Simulation ===
2 +==== Supervisor: María Pereira Martínez
3 3  ====
4 4  
5 -This project will explore new ways to communicate IGFAE’s strategic research areas—particle/astroparticle/nuclear physics, as well as Quantum IST, to young audiences. The goal will be to analyse the science communication & outreach on video formats, with some case studies (e.g. CERN, Fermilab, DESY, IFT, among others) to identify which formats, narrative structures and resources are more effective to communicate physics in an engaging way, whilst maintaining academic rigour.
6 -
7 -This analysis will be compared with the IGFAE’s social media activity, with the aim of proposing new ideas & approaches to help improve the Institute’s outreach activities in this area, in line with the 2025–2030 Communication Plan.
5 +Neutron tomography is an innovative technique for the detailed analysis of dense materials, allowing for the detection of internal defects and chemical composition without destroying the sample. Optimizing the main detector of the tomograph requires simulations that demand high computational resources. To address this, generative machine learning models offer a revolutionary approach to speed up simulations by learning to mimic the detector's response under different experimental conditions. In this project, the student will implement and train a generative neural network using Python libraries (like TensorFlow or PyTorch) to simulate the neutron tomograph and validate the model against traditional simulations.