Carl Case

Carl Case is a Senior Architect in Compute Architecture at NVIDIA, where he works on reduced-precision arithmetic for training deep neural networks. His focus is optimizing the entire stack of deep learning training, from hardware to high-level software, to accelerate the pace of AI development. Previously, he worked as a machine learning researcher on Deep Speech and its successor speech recognition systems at Baidu's Silicon Valley AI Lab.
Avatar photo

Posts by Carl Case

Simulation / Modeling / Design

NVIDIA Apex: Tools for Easy Mixed-Precision Training in PyTorch

Most deep learning frameworks, including PyTorch, train using 32-bit floating point (FP32) arithmetic by default. However, using FP32 for all operations is not... 8 MIN READ