Introduction
The course, consisting of two different modules, aims to provide an overview of the world of Deep Learning, closely linked to the enabling technology of supercomputing (HPC).
- The first module will provide a general overview of the main neural network architectures used in various application areas, combining theoretical explanations with practical exercise sessions to empower participants in the use of such algorithms.
- The second module will provide the basic knowledge for using an HPC machine in order to understand its potential within the application discussed in the first module. Subsequently, some use cases will be provided to motivate the use of HPC in the field of Deep Learning. In a practical session, several Deep Learning models will be implemented on an HPC infrastructure.
Target Audience
The training program is dedicated to individuals with a technical-scientific background, basic programming language skills, familiarity with the Linux operating system, and basic Machine Learning concepts. They should be interested in exploring and experimenting with advanced Deep Learning techniques using HPC.
Contents
- Different types of data
- Key issues related to data
- Neural network architectures (CNN, GAN, RNN, etc.)
- Basic concepts of HPC
Instructors
- Domitilla Brandoni
- TBD
Information and Accessibility
The course will take place over two days in person (duration: 12 hours). The dates are being finalized.
Course Program
Duration: 12 hours, spread over 2 days
Module 1
Objectives: The first module will focus on introducing the concepts related to Deep Learning and the main neural network architectures, as well as implementing the described models using the Python language.
Contents: (1) Introduction to key data issues; (2) Description of the architecture of some neural network models; (3) Practical exercises using the described models in Python.
Module 2
Objectives: The first part will provide some basic knowledge in the field of HPC. In a practical session, participants will learn how to interact with a supercomputer (login, data loading, code execution). The second part will describe some use cases to motivate the use of HPC in Deep Learning. In a practical session, several neural networks will be implemented on an HPC infrastructure.
Contents: (1) Introduction to HPC; (2) Practical exercises in an HPC environment; (3) Practical exercises on machine learning models in an HPC environment.
Are you interested in participating?
Please write to us at training_eurocc_italy@cineca.it.
