CNN Compression-Recovery Framework via Rank Allocation Decomposition With Knowledge Transfer

CNN Compression-Recovery Framework via Rank Allocation Decomposition With Knowledge Transfer

Low-rank decomposition is an effective way to decrease the model size of convolutional neural networks (CNNs). Nevertheless, selecting the layer-specific rank is a difficult task, because the layers are not equally redundant. The previous methods are mainly by manual, require expertise, or do not co...

Full description

Bibliographic Details
Main Authors: Zhonghong Ou, Yunfeng Liu, Huihui Kong, Meina Song, Yingxia Shao
Format: Article
Language:English
Published: IEEE 2019-01-01
Series:IEEE Access
Subjects:
Deep learning
neural network
decomposition
knowledge transfer
Online Access:https://ieeexplore.ieee.org/document/8786129/

Internet

https://ieeexplore.ieee.org/document/8786129/

Similar Items