Papers
Communities
Events
Blog
Pricing
Search
Open menu
Home
Papers
1912.00286
Cited By
Training Distributed Deep Recurrent Neural Networks with Mixed Precision on GPU Clusters
30 November 2019
Alexey Svyatkovskiy
J. Kates-Harbeck
W. Tang
Re-assign community
ArXiv
PDF
HTML
Papers citing
"Training Distributed Deep Recurrent Neural Networks with Mixed Precision on GPU Clusters"
2 / 2 papers shown
Title
RedBit: An End-to-End Flexible Framework for Evaluating the Accuracy of Quantized CNNs
A. M. Ribeiro-dos-Santos
João Dinis Ferreira
O. Mutlu
G. Falcão
MQ
21
1
0
15 Jan 2023
A Distributed Synchronous SGD Algorithm with Global Top-
k
k
k
Sparsification for Low Bandwidth Networks
S. Shi
Qiang-qiang Wang
Kaiyong Zhao
Zhenheng Tang
Yuxin Wang
Xiang Huang
Xiaowen Chu
40
135
0
14 Jan 2019
1