Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Ankur Agrawal

Accumulation Bit-Width Scaling For Ultra-Low Precision Training Of Deep Networks

Jan 19, 2019
Charbel Sakr, Naigang Wang, Chia-Yu Chen, Jungwook Choi, Ankur Agrawal, Naresh Shanbhag, Kailash Gopalakrishnan

* Published as a conference paper in ICLR 2019 

  Access Paper or Ask Questions

AdaComp : Adaptive Residual Gradient Compression for Data-Parallel Distributed Training

Dec 07, 2017
Chia-Yu Chen, Jungwook Choi, Daniel Brand, Ankur Agrawal, Wei Zhang, Kailash Gopalakrishnan

* IBM Research AI, 9 pages, 7 figures, AAAI18 accepted 

  Access Paper or Ask Questions

Deep Learning with Limited Numerical Precision

Feb 09, 2015
Suyog Gupta, Ankur Agrawal, Kailash Gopalakrishnan, Pritish Narayanan

* 10 pages, 6 figures, 1 table 

  Access Paper or Ask Questions