mlsanityData to 13 Oct 2026

CP-decomposition with Tensor Power Method for Convolutional Neural Networks Compression

arXiv 1701.07148 ↗First posted 25 Jan 2017cs.LGcited by 30

Not on the map. Why?

Abstract (arXiv)

Convolutional Neural Networks (CNNs) has shown a great success in many areas including complex image classification tasks. However, they need a lot of memory and computational cost, which hinders them from running in relatively low-end smart devices such as smart phones. We propose a CNN compression method based on CP-decomposition and Tensor Power Method. We also propose an iterative fine tuning, with which we fine-tune the whole network after decomposing each layer, but before decomposing the next layer. Significant reduction in memory and computation cost is achieved compared to state-of-the-art previous work with no more accuracy loss.

Key facts. Cited by 30 papers in the map (data through 13 Oct 2026). First posted on arXiv 25 Jan 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2021: 8; 2022: 3; 2023: 9; 2024: 5; 2025: 3; 2026: 2.

30 papers.

Sort:

By area:

  1. 2021Tensor Methods in Computer Vision and Deep Learningcited by 52
  2. 2023HALOC: Hardware-Aware Automatic Low-Rank Compression for Compact Neural Networkscited by 20
  3. 2023Tensor Decomposition for Model Reduction in Neural Networks: A Reviewcited by 14
  4. 2023A Survey on Approximate Edge AI for Energy Efficient Autonomous Driving Servicescited by 14
  5. 2023Geometry-aware training of factorized layers in tensor Tucker formatcited by 12
  6. 2021Low-rank Tensor Decomposition for Compression of Convolutional Neural Networks Using Funnel Regularizationcited by 6
  7. 2023Low Rank Optimization for Efficient Deep Learning: Making A Balance between Compact Architecture and Fast Trainingcited by 5
  8. 2025Tensorization is a powerful but underexplored tool for compression and interpretability of neural networkscited by 4
  9. 2022Position: Tensor Networks are a Valuable Asset for Green AIcited by 3
  10. 2022Canonical convolutional neural networkscited by 3
  11. 2024MIGS: Multi-Identity Gaussian Splatting via Tensor Decompositioncited by 3
  12. 2024Parameter-efficient Bayesian Neural Networks for Uncertainty-aware Depth Estimationcited by 3
18 more
  1. 2021Efficient Visual Recognition with Deep Neural Networks: A Survey on Recent Advances and New Directionscited by 2
  2. 2023Robust Adversarial Defense by Tensor Factorizationcited by 2
  3. 2023Boosting Defect Detection in Manufacturing using Tensor Convolutional Neural Networkscited by 2
  4. 2021Exploiting Elasticity in Tensor Ranks for Compressing Neural Networkscited by 1
  5. 2021Defensive Tensorizationcited by 1
  6. 2021A New Measure of Model Redundancy for Compressed Convolutional Neural Networkscited by 1
  7. 2022Design Automation for Fast, Lightweight, and Effective Deep Learning Models: A Surveycited by 1
  8. 2023When Parameter-efficient Tuning Meets General-purpose Vision-language Modelscited by 1
  9. 2021Nonlinear Tensor Ring Networkcited by 0
  10. 2021Selective Multiple Power Iteration: from Tensor PCA to gradient-based exploration of landscapescited by 0
  11. 2023Lightweight Attribute Localizing Models for Pedestrian Attribute Recognitioncited by 0
  12. 2024Convolutional Neural Network Compression via Dynamic Parameter Rank Pruningcited by 0
  13. 2024Reduced storage direct tensor ring decomposition for convolutional neural networks compressioncited by 0
  14. 2024Layer-Specific Optimization: Sensitivity Based Convolution Layers Basis Searchcited by 0
  15. 2025Efficient CNN Compression via Multi-method Low Rank Factorization and Feature Map Similaritycited by 0
  16. 2025Distribution-Aware Tensor Decomposition for Compression of Convolutional Neural Networkscited by 0
  17. 2026Optimizing Tensor Train Decomposition in DNNs for RISC-V Architectures Using Design Space Exploration and Compiler Optimizationscited by 0
  18. 2026Fast Tensorization of Neural Networks via Slice-wise Feature Distillationcited by 0