Welcome to Minerva Training Material Repository, the official archive collecting all resources from training events organized within the MINERVA project.
The goal of this repository is to provide a single shared reference point where event organizers can upload their presentations and make them accessible to all project stakeholders and the community.
MINERVA's consortium brings together unmatched expertise in HPC infrastructure, AI model development, and ethical innovation. Weβre committed to democratizing access to these technologies and driving European leadership.
Date | Training title | Organizers | Resources |
---|---|---|---|
16 Sep 2025 | HPC for AI Research - ICIAP 2025 | UNIMORE, CINECA | π PDF Folder | π Event Page |
<DATE> | <EVENT NAME> | <ORGANIZERS> | <LINKS TO RESOURCES> |
-
- Date: September 16, 2025
- Organizers: UNIMORE (University of Modena and Reggio Emilia), CINECA
- Slides: π PDF folder | π Event page
- Description This tutorial focuses on the computational aspects of training and deploying large-scale AI models using high-performance computing (HPC). Topics include distributed training, multi-GPU and multi-node acceleration, mixed-precision optimization, efficient data handling, and resource management on supercomputing clusters. Special attention is given to transformer-based architectures, self- and semi-supervised learning, and energy-efficient AI practices. Participants will gain both theoretical insights and hands-on experience with frameworks such as PyTorch and DeepSpeed, learning best practices for scalable and sustainable AI research on HPC infrastructures.
-
- Date: <DATE>
- Organizers: <ORGANIZER>
- Slides: π PDF folder | π Event page
- Description <DESCRIPTION>
If you are organizing a training event and want to upload the slides:
- Create (locally) a folder named
YYYY-MM-DD_EventName
inside thepdfs
directory. - Add all related PDF files (slides / training materials).
- Commit and open a pull request (PR) to this repository, including:
- A short description of the event (date, organizer, short abstract, event webpage).
- The name of the folder / directory and list of uploaded PDF files.
All training materials (PDFs) remain the property of their respective authors / organizers.
This repository serves as a shared archive with educational and collaborative purposes within the MINERVA project.
Any external use should be agreed with the original authors (e.g. citation, redistribution).