Speaker: Andrea Pilzer, Ph.D. (NVIDIA AI Technology Center, Italy)
Slides: https://hpc.fau.de/files/2026/01/20260116_FAU.pdf
Abstract:
In this talk, we explore scaling laws to understand the rationale behind large-scale training. We discuss how to apply parallelization techniques effectively, ensuring they are used where they deliver the greatest benefit. Finally, we introduce low-precision training methods to maximize cluster performance and efficiency.
For a list of past and upcoming NHR PerfLab seminar events, please see: https://hpc.fau.de/research/nhr-perflab-seminar-series/
