mlsanityData to 14 Oct 2026

Pruning Ternary Quantization

arXiv 2107.10998 ↗First posted 23 Jul 2021cs.CV cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Inference time, model size, and accuracy are three key factors in deep model compression. Most of the existing work addresses these three key factors separately as it is difficult to optimize them all at the same time. For example, low-bit quantization aims at obtaining a faster model; weight sharing quantization aims at improving compression ratio and accuracy; and mixed-precision quantization aims at balancing accuracy and inference time. To simultaneously optimize bit-width, model size, and accuracy, we propose pruning ternary quantization (PTQ): a simple, effective, symmetric ternary quantization method. We integrate L2 normalization, pruning, and the weight decay term to reduce the weight discrepancy in the gradient estimator during quantization, thus producing highly compressed ternary weights. Our method brings the highest test accuracy and the highest compression ratio. For example, it produces a 939kb (49×) 2bit ternary ResNet-18 model with only 4% accuracy drop on the ImageNet dataset. It compresses 170MB Mask R-CNN to 5MB (34×) with only 2.8% average precision drop. Our method is verified on image classification, object detection/segmentation tasks with different network structures such as ResNet-18, ResNet-50, and MobileNetV2.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 42 of its 59 references that are in the map. First posted on arXiv 23 Jul 2021.

Builds on

42 of 59 references.

References · 42

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  3. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
  4. 2017Mask R-CNNcited by 7,277
  5. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 5,045
  6. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,795
  7. 2015Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Codingcited by 2,329
  8. 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,079
  9. 2016Pruning Filters for Efficient ConvNetscited by 1,143
  10. 2016XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networkscited by 849
  11. 2017SphereFace: Deep Hypersphere Embedding for Face Recognitioncited by 737
  12. 2017Channel Pruning for Accelerating Very Deep Neural Networkscited by 702
30 more
  1. 2019Learned Step Size Quantizationcited by 628
  2. 2015BinaryConnect: Training Deep Neural Networks with binary weights during propagationscited by 592
  3. 2016DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradientscited by 569
  4. 2016Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networkscited by 543
  5. 2021BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstructioncited by 479
  6. 2018PACT: Parameterized Clipping Activation for Quantized Neural Networkscited by 467
  7. 2018HAQ: Hardware-Aware Automated Quantization with Mixed Precisioncited by 443
  8. 2019HAWQ: Hessian AWare Quantization of Neural Networks with Mixed-Precisioncited by 402
  9. 2019Q-BERT: Hessian Based Ultra Low Precision Quantization of BERTcited by 359
  10. 2016EIE: Efficient Inference Engine on Compressed Deep Neural Networkcited by 295
  11. 2018LQ-Nets: Learned Quantization for Highly Accurate and Compact Deep Neural Networkscited by 274
  12. 2020ZeroQ: A Novel Zero Shot Quantization Frameworkcited by 269
  13. 2017Variational Dropout Sparsifies Deep Neural Networkscited by 265
  14. 2016Network Trimming: A Data-Driven Neuron Pruning Approach towards Efficient Deep Architecturescited by 221
  15. 2015Quantized Convolutional Neural Networks for Mobile Devicescited by 215
  16. 2016Trained Ternary Quantizationcited by 208
  17. 2016Ternary Weight Networkscited by 204
  18. 2014Compressing Deep Convolutional Networks using Vector Quantizationcited by 195
  19. 2017Towards Accurate Binary Convolutional Neural Networkcited by 149
  20. 2019Additive Powers-of-Two Quantization: An Efficient Non-uniform Discretization for Neural Networkscited by 148
  21. 2020Training with Quantization Noise for Extreme Model Compressioncited by 145
  22. 2018Mixed Precision Quantization of ConvNets via Differentiable Neural Architecture Searchcited by 125
  23. 2019And the Bit Goes Down: Revisiting the Quantization of Neural Networkscited by 78
  24. 2020An Overview of Neural Network Compressioncited by 59
  25. 2017Compression-aware Training of Deep Networkscited by 46
  26. 2017Learning Discrete Weights Using the Local Reparameterization Trickcited by 29
  27. 2020Permute, Quantize, and Fine-tune: Efficient Compression of Neural Networkscited by 28
  28. 2018Training wide residual networks for deployment using a single bit for each weightcited by 10
  29. 2017Model compression as constrained optimization, with application to neural nets. Part II: quantizationcited by 8
  30. 2020FATNN: Fast and Accurate Ternary Neural Networkscited by 7

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2023Ternary Quantization: A Survey
  2. 2023Designing strong baselines for ternary neural network quantization through support and mass equalization
  3. 2026Joint Structural Pruning and Mixed-Precision Quantization for LLM Compression
  4. 2022Hyperspherical Quantization: Toward Smaller and More Accurate Models
  5. 2023Towards Optimal Compression: Joint Pruning and Quantization
  6. 2025Integrating Pruning with Quantization for Efficient Deep Neural Networks Compression
  7. 2025CoDeQ: End-to-End Joint Model Compression with Dead-Zone Quantizer for High-Sparsity and Low-Precision Networks
  8. 2026CAT-Q: Cost-efficient and Accurate Ternary Quantization for LLMs