Search Results for author:
Found 1 papers, 0 papers with code
Natural Compression for Distributed Deep Learning
Our technique is applied individually to all entries of the to-be-compressed update vector and works by randomized rounding to the nearest (negative or positive) power of two, which can be computed in a "natural" way by ignoring the mantissa.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
