This day is full; the Workshop will be repeated on 28th October, so please try to register under

https://indico.scc.kit.edu/event/668/overview

Content of the Workshop

The goal of this workshop is to convey advanced knowledge in the area of machine learning and its practical application. In this context, practical exercises are solved with IBM PowerAI. The Power9 architecture has the advantage of increased memory bandwidth compared with the X86 one and allows for a high-performance connection to NVIDIA V100 GPUs via NVLink 2.0.

Concepts examined in the workshop include:

• Data Preparation

• Hardware Acceleration

  • Distributed Deep Learning

  • Large Model Support

  • Elastic Distributed Training

• Auto Machine Learning (H2O Driverless AI)

• Auto Deep Learning (PowerAI Vision)

• Data Science in Enterprise Environments (processes, roles and software)

Which previous knowledge might be helpful?

• basics of programming, ideally python

• rudimentary mathematical and statistical knowledge

Getting to the Venue

The workshop will be held at Campus North in the SCC building 449. The best way to get there, is to take the KIT Shuttle, which starts at 08:00 or 08:30 at Campus South and stay until KIT-CN H5 "Karlsruher Allee". We will open the room in time for the people arriving with the earlier shuttle, but the workshop will really get going only, when the 08:30 shuttle arrives.