mlsanityData to 7 Oct 2026

Network Pruning via Feature Shift Minimization

arXiv 2207.02632 ↗First posted 6 Jul 2022cs.CVcited by 4

Not on the map. Why?

Abstract (arXiv)

Channel pruning is widely used to reduce the complexity of deep network models. Recent pruning methods usually identify which parts of the network to discard by proposing a channel importance criterion. However, recent studies have shown that these criteria do not work well in all conditions. In this paper, we propose a novel Feature Shift Minimization (FSM) method to compress CNN models, which evaluates the feature shift by converging the information of both features and filters. Specifically, we first investigate the compression efficiency with some prevalent methods in different layer-depths and then propose the feature shift concept. Then, we introduce an approximation method to estimate the magnitude of the feature shift, since it is difficult to compute it directly. Besides, we present a distribution-optimization algorithm to compensate for the accuracy loss and improve the network compression efficiency. The proposed method yields state-of-the-art performance on various benchmark networks and datasets, verified by extensive experiments. Our codes are available at: https://github.com/lscgx/FSM.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 40 of its 52 references that are in the map. First posted on arXiv 6 Jul 2022.

Builds on

40 of 52 references.

References · 40

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  3. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  4. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  5. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  6. 2016Densely Connected Convolutional Networkscited by 6,691
  7. 2014Fully Convolutional Networks for Semantic Segmentationcited by 4,845
  8. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
  9. 2014Going Deeper with Convolutionscited by 4,231
  10. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,827
  11. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,590
  12. 2016DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFscited by 3,239
28 more
  1. 2018Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentationcited by 3,210
  2. 2013Rich feature hierarchies for accurate object detection and semantic segmentationcited by 2,160
  3. 2015Learning both Weights and Connections for Efficient Neural Networkscited by 1,998
  4. 2016Pruning Filters for Efficient ConvNetscited by 1,129
  5. 2017Channel Pruning for Accelerating Very Deep Neural Networkscited by 697
  6. 2017Learning Efficient Convolutional Networks through Network Slimmingcited by 673
  7. 2017ThiNet: A Filter Level Pruning Method for Deep Neural Network Compressioncited by 438
  8. 2019BMN: Boundary-Matching Network for Temporal Action Proposal Generationcited by 327
  9. 2018Soft Filter Pruning for Accelerating Deep Convolutional Neural Networkscited by 313
  10. 2020HRank: Filter Pruning using High-Rank Feature Mapcited by 291
  11. 2019MetaPruning: Meta Learning for Automatic Neural Network Channel Pruningcited by 226
  12. 2020ReActNet: Towards Precise Binary Neural Network with Generalized Activation Functionscited by 183
  13. 2017Data-Driven Sparse Structure Selection for Deep Neural Networkscited by 181
  14. 2018Discrimination-aware Channel Pruning for Deep Neural Networkscited by 171
  15. 2019Towards Optimal Structured CNN Pruning via Generative Adversarial Learningcited by 145
  16. 2019Gate Decorator: Global Filter Pruning Method for Accelerating Deep Convolutional Neural Networkscited by 118
  17. 2020Neural Pruning via Growing Regularizationcited by 97
  18. 2019Towards Efficient Model Compression via Learned Global Rankingcited by 75
  19. 2020Operation-Aware Soft Channel Pruning using Differentiable Maskscited by 74
  20. 2021Convolutional Neural Network Pruning with Structural Redundancy Reductioncited by 73
  21. 2020DHP: Differentiable Meta Pruning via HyperNetworkscited by 57
  22. 2021Towards Compact CNNs via Collaborative Compressioncited by 36
  23. 2020Hit-Detector: Hierarchical Trinity Architecture Search for Object Detectioncited by 32
  24. 2018Leveraging Filter Correlations for Deep Model Compressioncited by 23
  25. 2019DistInit: Learning Video Representations Without a Single Labeled Videocited by 21
  26. 2017Predicting Video Saliency with Object-to-Motion CNN and Two-layer Convolutional LSTMcited by 16
  27. 2021Model Pruning Based on Quantified Similarity of Feature Mapscited by 6
  28. 2021Batch Normalization Tells You Which Filter is Importantcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 3; 2026: 1.

4 papers.

  1. 2025Enhanced Structured Lasso Pruning with Class-wise Informationcited by 1
  2. 2025LinDeps: A Fine-tuning Free Post-Pruning Method to Remove Layer-Wise Linear Dependencies with Guaranteed Performance Preservationcited by 1
  3. 2025IPPRO: Importance-based Pruning with PRojective Offset for Magnitude-indifferent Structural Pruningcited by 0
  4. 2026DCP-Prune: Ultra-Low Token Pruning with Distribution Consistency Preservationcited by 0