Decoupled neural network training with re-computation and weight prediction
To break the three lockings during backpropagation (BP) process for neural network training, multiple decoupled learning methods have been investigated recently. These methods either lead to significant drop in accuracy performance or suffer from dramatic increase in memory usage. In this paper, a n...
Saved in:
Main Authors: | Peng, Jiawei, Xu, Yicheng, Lin, Zhiping, Weng, Zhenyu, Yang, Zishuo, Zhuang, Huiping |
---|---|
Other Authors: | School of Electrical and Electronic Engineering |
Format: | Article |
Language: | English |
Published: |
2023
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/169712 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Similar Items
-
A deep-learning approach to the dynamics of Landau-Zener transitions
by: Gao, Linliang, et al.
Published: (2021) -
Accumulated decoupled learning with gradient staleness mitigation for convolutional neural networks
by: Zhuang, Huiping, et al.
Published: (2024) -
Attention multihop graph and multiscale convolutional fusion network for hyperspectral image classification
by: Zhou, Hao, et al.
Published: (2023) -
A neural-network-based-method on speed control of ultrasonic motors
by: Liang, Y.C., et al.
Published: (2014) -
A neural network approach to determining optimal inspection sampling size for CMM
by: Zhang, Y.F., et al.
Published: (2014)