Booking required
Booking deadline: Thursday 5 March 2020

Today the computational requirements of deep neural networks that enable complex AI applications like autonomous driving, language translation, and speech synthesis are enormous. A single training cycle can take weeks on a single GPU or even years for the larger datasets like those used in self-driving car research. Using multiple GPUs for deep learning can significantly shorten the time required to train lots of data, making solving complex problems with deep learning feasible.

At the University of Luxembourg we have access to state-of-the-art HPC systems perfectly suitable for the largest Deep Learning Training sessions. This course will give you all the necessary tools to utilize that power using just Python and the framework of choice between TensorFlow, PyTorch, Caffe2, MXNET, CNTK.

This course will teach you how to use multiple GPUs to training neural networks. You'll learn:

Approaches to multi-GPU training
Algorithmic and engineering challenges to large-scale training
Key techniques used to overcome the challenges mentioned above
Upon completion, you'll be able to effectively parallelize training of deep neural networks using TensorFlow.

This workshop is hands-on with multiple Python coding exercises, thus beginner's level Python experience will be beneficial. Upon successful completion of the assessment, participants will receive an NVIDIA DLI certificate to recognize their subject matter competency and support professional career growth.

This course is the premier course on deep learning deployment offered by the NVIDIA Deep Learning Institute. You can find more detailed information at:
https://www.nvidia.com/content/dam/en-zz/Solutions/deep-learning/deep-learning-education/Fundamentals_of_Deep_Learning_for_Multi-GPUs.pdf

Note that this course requires a minimum participation of 20. It is prioritized to SnT but soon registrations will open to all university members.

Photography

Please note that a photographer will be present at this event. The images may be published by the University, e.g. on its social media, its website and in print products. If you do not wish to be photographed, please alert the organisers and the photographer.

Data processing policy

The Interdisciplinary Centre for Security, Reliability and Trust (hereafter “SnT”) collects and processes your personal data in the framework of your registration and your participation in this event and the communication of its future activities. SnT processes and keeps your personal information only to the extent necessary to fulfill the purpose stated.

During the registration process your personal data is being collected and processed on the Eventbrite website by Eventbrite Inc. located in the United States. This will involve the transfer of personal data to countries that do not offer the same level of protection as the European Union regarding data protection rules. Eventbrite Inc. participates in and complies with the EU-U.S. Privacy Shield framework as set forth by the US Department of Commerce regarding the collection, use and retention of personal data. Eventbrite may process your personal data as a controller for different purposes than the purposes of the University of Luxembourg. More information is provided on Eventbrite Inc.’s website in their EU-US Privacy Shield Notice, Privacy Notice and Terms of Services.

You can deregister from the event directly on the Eventbrite website and withdraw your consent for the processing of personal data by SnT for the purpose stated above. You have the right to access, rectify, and erase your personal data as well as further rights described on the University of Luxembourg's website. You can exercise your rights by following the procedure described on the website.

NVIDIA Workshop: Fundamentals of Deep Learning for Multi-GPUs