mlsanityData to 7 Oct 2026

A Multi-objective Complex Network Pruning Framework Based on Divide-and-conquer and Global Performance Impairment Ranking

arXiv 2303.16212 ↗First posted 28 Mar 2023cs.LG cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Model compression plays a vital role in the practical deployment of deep neural networks (DNNs), and evolutionary multi-objective (EMO) pruning is an essential tool in balancing the compression rate and performance of the DNNs. However, due to its population-based nature, EMO pruning suffers from the complex optimization space and the resource-intensive structure verification process, especially in complex networks. To this end, a multi-objective complex network pruning framework based on divide-and-conquer and global performance impairment ranking (EMO-DIR) is proposed in this paper. Firstly, a divide-and-conquer EMO network pruning method is proposed, which decomposes the complex task of EMO pruning on the entire network into easier sub-tasks on multiple sub-networks. On the one hand, this decomposition narrows the pruning optimization space and decreases the optimization difficulty; on the other hand, the smaller network structure converges faster, so the proposed algorithm consumes lower computational resources. Secondly, a sub-network training method based on cross-network constraints is designed, which could bridge independent EMO pruning sub-tasks, allowing them to collaborate better and improving the overall performance of the pruned network. Finally, a multiple sub-networks joint pruning method based on EMO is proposed. This method combines the Pareto Fronts from EMO pruning results on multiple sub-networks through global performance impairment ranking to design a joint pruning scheme. The rich experiments on CIFAR-10/100 and ImageNet-100/1k are conducted. The proposed algorithm achieves a comparable performance with the state-of-the-art pruning methods.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 28 of its 53 references that are in the map. First posted on arXiv 28 Mar 2023.

Builds on

28 of 53 references.

References · 28

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  4. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  5. 2014Going Deeper with Convolutionscited by 4,231
  6. 2015Learning both Weights and Connections for Efficient Neural Networkscited by 1,998
  7. 2014How transferable are features in deep neural networks?cited by 1,216
  8. 2016Pruning Filters for Efficient ConvNetscited by 1,129
  9. 2017Channel Pruning for Accelerating Very Deep Neural Networkscited by 697
  10. 2018Filter Pruning via Geometric Median for Deep Convolutional Neural Networks Accelerationcited by 365
  11. 2018Soft Filter Pruning for Accelerating Deep Convolutional Neural Networkscited by 313
  12. 2020HRank: Filter Pruning using High-Rank Feature Mapcited by 291
16 more
  1. 2020ZeroQ: A Novel Zero Shot Quantization Frameworkcited by 265
  2. 2016Dynamic Network Surgery for Efficient DNNscited by 233
  3. 2017NISP: Pruning Networks using Neuron Importance Score Propagationcited by 228
  4. 2017Data-Driven Sparse Structure Selection for Deep Neural Networkscited by 181
  5. 2019Towards Optimal Structured CNN Pruning via Generative Adversarial Learningcited by 145
  6. 2022Deep Hierarchical Semantic Segmentationcited by 84
  7. 2019Network Pruning via Transformable Architecture Searchcited by 80
  8. 2017More is Less: A More Complicated Network with Less Inference Complexitycited by 66
  9. 2021CTNet: Context-based Tandem Network for Semantic Segmentationcited by 43
  10. 2021Student Network Learning via Evolutionary Knowledge Distillationcited by 30
  11. 2018Asymptotic Soft Filter Pruning for Deep Convolutional Neural Networkscited by 27
  12. 2020Filter Sketch for Network Pruningcited by 20
  13. 2021SiMaN: Sign-to-Magnitude Network Binarizationcited by 18
  14. 2021Dynamic Attention guided Multi-Trajectory Analysis for Single Object Trackingcited by 17
  15. 2022Learning Weakly-Supervised Contrastive Representationscited by 6
  16. 2018Building Efficient ConvNets using Redundant Feature Pruningcited by 5

Built on by

Not cited yet.