Reversible designs for extreme memory cost reduction of CNN training

Reversible designs for extreme memory cost reduction of CNN training

Abstract Training Convolutional Neural Networks (CNN) is a resource-intensive task that requires specialized hardware for efficient computation. One of the most limiting bottlenecks of CNN training is the memory cost associated with storing the activation values of hidden layers. These values are ne...

Full description

Bibliographic Details
Main Authors: Tristan Hascoet, Quentin Febvre, Weihao Zhuang, Yasuo Ariki, Tetsuya Takiguchi
Format: Article
Language:English
Published: SpringerOpen 2023-01-01
Series:EURASIP Journal on Image and Video Processing
Subjects:
CNN
Reversibility
Memory optimization
Numerical analysis
Online Access:https://doi.org/10.1186/s13640-022-00601-w

Internet

https://doi.org/10.1186/s13640-022-00601-w

Similar Items