Hands-on Introduction to Deep Learning with PyTorch

Course Description

This course offers a concise introduction to the fundamental concepts of deep learning, covering neural network basics, training methods, and specific applications such as convolutional neural networks for computer vision and transformer models for natural language processing.

More specifically, the following topics will be covered:

- Fundamentals of neural networks
- Training deep learning models: stochastic gradient descent, optimizers, loss functions, regularization, etc.
- Training on multiple GPUs
- Convolutional Neural Networks (CNNs) for computer vision
- Natural Language Processing (NLP) with transformers: attention mechanisms, transformer architecture, etc.

The lessons will combine theoretical content with hands-on practice using PyTorch. Practical exercises will be run on the Piz Daint supercomputer at CSCS.

Target audience

This workshop is designed for researchers who are beginning to use deep learning in their work and are seeking a fast-paced introduction to the field. The goal is to provide foundational knowledge and guidance so that participants can continue their deep learning journey independently.

While PyTorch will be used during the practical sessions and some of its concepts will be introduced, this course is not a full introduction to PyTorch. The primary focus is on deep learning concepts. Participants are expected to be familiar with NumPy, or equivalent numerical computation and linear algebra packages in Python.

Agenda

The course will begin on Wednesday, July 2, 2025 at 10:00 and conclude on Friday, July 4 at 16:30 (Central European Summer Time, CEST). A detailed agenda will be shared closer to the event date.

Instructors

Dr. Rocco Meli (Research Software Engineer, CSCS)
Dr. Rafael Sarmiento (HPC Application Engineer, CSCS)
Dr. Fabian Bösch (Software Engineer, CSCS)

Participation Fee and Registration

All participants must register for the course. The registration fee includes lunch and coffee breaks.

Course Fee: CHF 240.00

Deadline for registration: Monday, June 23, 2025

Kindly note that the course can take place only if there are sufficient confirmed registrations received by the deadline. The minimum number of participants is eight. Registration for the course will automatically close when we reach the maximum number of participants (20). If the minimum number of participants is not reached, the in-person event will be cancelled.

Inquiries

Inquiries may be addressed to Rafael Sarmiento (rafael.sarmiento@cscs.ch).

Directions

Kindly note that no parking space is available at the Swiss National Supercomputing Centre. There is a Park & Ride Resega parking lot, within five minutes walk from CSCS.You are encouraged to travel by public transportation. The closest bus stop to CSCS is "Lugano, Stadio." From Lugano railway station, you should take bus number 4 or 6. Suggestions regarding travels and accommodation are available here.