mlsanityData to 14 Oct 2026

Performance-aware Approximation of Global Channel Pruning for Multitask CNNs

arXiv 2303.11923 ↗First posted 21 Mar 2023cs.CVcited by 9

Not on the map. Why?

Abstract (arXiv)

Global channel pruning (GCP) aims to remove a subset of channels (filters) across different layers from a deep model without hurting the performance. Previous works focus on either single task model pruning or simply adapting it to multitask scenario, and still face the following problems when handling multitask pruning: 1) Due to the task mismatch, a well-pruned backbone for classification task focuses on preserving filters that can extract category-sensitive information, causing filters that may be useful for other tasks to be pruned during the backbone pruning stage; 2) For multitask predictions, different filters within or between layers are more closely related and interacted than that for single task prediction, making multitask pruning more difficult. Therefore, aiming at multitask model compression, we propose a Performance-Aware Global Channel Pruning (PAGCP) framework. We first theoretically present the objective for achieving superior GCP, by considering the joint saliency of filters from intra- and inter-layers. Then a sequentially greedy pruning strategy is proposed to optimize the objective, where a performance-aware oracle criterion is developed to evaluate sensitivity of filters to each task and preserve the globally most task-related filters. Experiments on several multitask datasets show that the proposed PAGCP can reduce the FLOPs and parameters by over 60% with minor performance drop, and achieves 1.2x~3.3x acceleration on both cloud and mobile platforms.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Builds on 66 of its 79 references that are in the map. First posted on arXiv 21 Mar 2023.

Builds on

66 of 79 references.

References · 66

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  3. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
  4. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,733
  5. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,442
  6. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,852
  7. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,845
  8. 2015SSD: Single Shot MultiBox Detectorcited by 3,130
  9. 2015Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Codingcited by 2,329
  10. 2018The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networkscited by 2,262
  11. 2014FitNets: Hints for Thin Deep Netscited by 1,653
  12. 2019Deep High-Resolution Representation Learning for Visual Recognitioncited by 1,420
38 more
  1. 2019Objects as Pointscited by 1,175
  2. 2016Pruning Filters for Efficient ConvNetscited by 1,143
  3. 2021YOLOX: Exceeding YOLO Series in 2021cited by 1,071
  4. 2019EfficientDet: Scalable and Efficient Object Detectioncited by 1,020
  5. 2014Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognitioncited by 720
  6. 2017Channel Pruning for Accelerating Very Deep Neural Networkscited by 702
  7. 2017Learning Efficient Convolutional Networks through Network Slimmingcited by 676
  8. 2018Rethinking the Value of Network Pruningcited by 599
  9. 2019Importance Estimation for Neural Network Pruningcited by 541
  10. 2018End-to-End Multi-Task Learning with Attentioncited by 534
  11. 2016Cross-stitch Networks for Multi-task Learningcited by 467
  12. 2018AMC: AutoML for Model Compression and Acceleration on Mobile Devicescited by 378
  13. 2018Filter Pruning via Geometric Median for Deep Convolutional Neural Networks Accelerationcited by 367
  14. 2018Soft Filter Pruning for Accelerating Deep Convolutional Neural Networkscited by 315
  15. 2020HRank: Filter Pruning using High-Rank Feature Mapcited by 292
  16. 2018Model compression via distillation and quantizationcited by 289
  17. 2014Detect What You Can: Detecting and Representing Objects using Holistic Models and Body Partscited by 232
  18. 2017NISP: Pruning Networks using Neuron Importance Score Propagationcited by 228
  19. 2019MetaPruning: Meta Learning for Automatic Neural Network Channel Pruningcited by 227
  20. 2018PAD-Net: Multi-Tasks Guided Prediction-and-Distillation Network for Simultaneous Depth Estimation and Scene Parsingcited by 224
  21. 2016Network Trimming: A Data-Driven Neuron Pruning Approach towards Efficient Deep Architecturescited by 221
  22. 2015Quantized Convolutional Neural Networks for Mobile Devicescited by 215
  23. 2020Scaled-YOLOv4: Scaling Cross Stage Partial Networkcited by 211
  24. 2019Distilling Object Detectors with Fine-grained Feature Imitationcited by 189
  25. 2018Discrimination-aware Channel Pruning for Deep Neural Networkscited by 171
  26. 2020MTI-Net: Multi-Scale Task Interaction Networks for Multi-Task Learningcited by 157
  27. 2019Towards Optimal Structured CNN Pruning via Generative Adversarial Learningcited by 146
  28. 2019Fair DARTS: Eliminating Unfair Advantages in Differentiable Architecture Searchcited by 131
  29. 2017Single-Shot Refinement Neural Network for Object Detectioncited by 129
  30. 2018NDDR-CNN: Layerwise Feature Fusing in Multi-Task CNNs by Neural Discriminative Dimensionality Reductioncited by 127
  31. 2019Provable Filter Pruning for Efficient Neural Networkscited by 98
  32. 2020SCOP: Scientific Control for Reliable Neural Network Pruningcited by 93
  33. 2020Channel Pruning via Automatic Structure Searchcited by 85
  34. 2020ResRep: Lossless CNN Pruning via Decoupling Remembering and Forgettingcited by 85
  35. 2020DARTS-: Robustly Stepping out of Performance Collapse Without Indicatorscited by 81
  36. 2019Network Pruning via Transformable Architecture Searchcited by 80
  37. 2020EagleEye: Fast Sub-net Evaluation for Efficient Neural Network Pruningcited by 80
  38. 2021CHIP: CHannel Independence-based Pruning for Compact Neural Networkscited by 78

50 most cited of 66 All 66 (JSON)

Built on by

Cited by year

Citing papers per year: 2023: 1; 2024: 4; 2025: 4.

9 papers.

  1. 2024FedLPS: Heterogeneous Federated Learning for Multiple Tasks with Local Parameter Sharingcited by 8
  2. 2024Once for Both: Single Stage of Importance and Sparsity Search for Vision Transformer Compressioncited by 8
  3. 2024Enhanced Sparsification via Stimulative Trainingcited by 2
  4. 2024AdapMTL: Adaptive Pruning Framework for Multitask Learning Modelcited by 2
  5. 2025One-Cycle Structured Pruning via Stability-Driven Subnetwork Searchcited by 1
  6. 2023Beta-Rank: A Robust Convolutional Filter Pruning Method For Imbalanced Medical Image Analysiscited by 0
  7. 2025Loss-Aware Automatic Selection of Structured Pruning Criteria for Deep Neural Network Accelerationcited by 0
  8. 2025MGAA: Multi-Granular Adaptive Allocation fof Low-Rank Compression of LLMscited by 0
  9. 2025IPPRO: Importance-based Pruning with PRojective Offset for Magnitude-indifferent Structural Pruningcited by 0