Progressive Network Grafting With Local Features Embedding for Few-Shot Knowledge Distillation

Progressive Network Grafting With Local Features Embedding for Few-Shot Knowledge Distillation

Compared with traditional knowledge distillation, which relies on a large amount of data, few-shot knowledge distillation can distill student networks with good performance using only a small number of samples. Some recent studies treat the network as a combination of a series of network blocks, ado...

Full description

Bibliographic Details
Main Author: Weidong Du
Format: Article
Language:English
Published: IEEE 2022-01-01
Series:IEEE Access
Subjects:
Knowledge distillation
few-shot learning
model compression
features embedding
Online Access:https://ieeexplore.ieee.org/document/9934906/

Internet

https://ieeexplore.ieee.org/document/9934906/

Similar Items