Search Results for author: Nuno Roma

Found 1 papers, 1 papers with code

PositNN: Training Deep Neural Networks with Mixed Low-Precision Posit

1 code implementation30 Apr 2021 Gonçalo Raposo, Pedro Tomás, Nuno Roma

Low-precision formats have proven to be an efficient way to reduce not only the memory footprint but also the hardware resources and power consumption of deep learning computations.

Cannot find the paper you are looking for? You can Submit a new open access paper.