Combining Global Sparse Gradients with Local Gradients in Distributed Neural Network Training
9139Knowledge Triples
| Source Entity | Relationship | Target Entity |
|---|---|---|
|
Gradient Compression
Method |
usedFor | |
|
Stochastic Gradient Descent
Method |
enhancedBy |
Gradient Compression
Method |
|
Transformer Models
Method |
affects |
Model Performance
Metric |