Filter Grafting for Deep Neural Networks

@article{Meng2020FilterGF,
  title={Filter Grafting for Deep Neural Networks},
  author={Fanxu Meng and H. Cheng and Ke Li and Zhixin Xu and Rongrong Ji and Xing Sun and G. Lu},
  journal={2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2020},
  pages={6598-6606}
}
  • Fanxu Meng, H. Cheng, +4 authors G. Lu
  • Published 2020
  • Computer Science
  • 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
This paper proposes a new learning paradigm called filter grafting, which aims to improve the representation capability of Deep Neural Networks (DNNs). The motivation is that DNNs have unimportant (invalid) filters (e.g., l1 norm close to 0). These filters limit the potential of DNNs since they are identified as having little effect on the network. While filter pruning removes these invalid filters for efficiency consideration, filter grafting re-activates them from an accuracy boosting… Expand
8 Citations
CondenseNet V2: Sparse Feature Reactivation for Deep Networks
  • Le Yang, Haojun Jiang, +4 authors Qi Tian
  • Computer Science
  • ArXiv
  • 2021
  • PDF
Progressive Network Grafting for Few-Shot Knowledge Distillation
  • 2
  • PDF
Compacting Deep Neural Networks for Internet of Things: Methods and Applications
  • Ke Zhang, Hanbo Ying, +4 authors Hongfang Yu
  • Computer Science
  • ArXiv
  • 2021
  • PDF
Mutual Contrastive Learning for Visual Representation Learning
  • PDF
Ferrograph image classification
  • 1
  • PDF
Vehicle Re-Identification Based on Complementary Features
  • 1
  • PDF

References

SHOWING 1-10 OF 33 REFERENCES
Soft Filter Pruning for Accelerating Deep Convolutional Neural Networks
  • 296
  • PDF
RePr: Improved Training of Convolutional Filters
  • 29
  • Highly Influential
  • PDF
Opening the Black Box of Deep Neural Networks via Information
  • 701
  • PDF
Utilizing Information Bottleneck to Evaluate the Capability of Deep Neural Networks for Image Classification †
  • 7
  • PDF
Filter Pruning via Geometric Median for Deep Convolutional Neural Networks Acceleration
  • 249
  • PDF
Importance Estimation for Neural Network Pruning
  • 144
  • PDF
Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification
  • 9,629
  • Highly Influential
  • PDF
Deep Residual Learning for Image Recognition
  • 65,615
  • PDF
...
1
2
3
4
...