Patrick Donnelly and Gary Burnett

We explain how NVIDIA GPUs accelerate computation in two common areas of research: computer vision and natural language processing. We discuss advances in neural network architecture from NVIDIA Research, introduce NVIDIA libraries for training and deploying neural networks (cuDNN and TensorRT), and provide a live demo of training and inference using NVIDIA GPU Cloud.
Patrick Donnelly and Gary Burnett
NVIDIA Corporation
Wed, 05/09/2018 - 1:00pm
RC-4082, The Red Centre, UNSW