DSP decouples forward and backward passes in model-parallel deep CNN training by giving each layer block a preset staleness, yielding speedups up to 4.8x with comparable or better accuracy.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
On the Acceleration of Deep Learning Model Parallelism with Staleness
DSP decouples forward and backward passes in model-parallel deep CNN training by giving each layer block a preset staleness, yielding speedups up to 4.8x with comparable or better accuracy.