Fully decoupled neural network learning using delayed gradients
Training neural networks with back-propagation (BP) requires a sequential passing of activations and gradients. This has been recognized as the lockings (i.e., the forward, backward, and update lockings) among modules (each module contains a stack of layers) inherited from the BP. In this paper,...
Saved in:
Main Authors: | Zhuang, Huiping, Wang, Yi, Liu, Qinglai, Lin, Zhiping |
---|---|
Other Authors: | School of Electrical and Electronic Engineering |
Format: | Article |
Language: | English |
Published: |
2024
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/174476 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Similar Items
-
Accumulated decoupled learning with gradient staleness mitigation for convolutional neural networks
by: Zhuang, Huiping, et al.
Published: (2024) -
Decoupling Smith predictor design for multivariable systems with multiple time delays
by: Wang, Q.-G., et al.
Published: (2014) -
Decoupling internal model control for multivariable systems with multiple time delays
by: Wang, Q.-G., et al.
Published: (2014) -
Decoupling internal model control for multivariable systems with multiple time delays
by: Wang, Q.-G., et al.
Published: (2014) -
High efficient crossing-order decoupling in Vold-Kalman filtering order tracking based on independent component analysis
by: Guo, Y., et al.
Published: (2014)