QFT: Post-training quantization via fast joint finetuning of all degrees of freedom
Not on the map. Why?
Abstract (arXiv)
The post-training quantization (PTQ) challenge of bringing quantized neural net accuracy close to original has drawn much attention driven by industry demand. Many of the methods emphasize optimization of a specific degree-of-freedom (DoF), such as quantization step size, preconditioning factors, bias fixing, often chained to others in multi-step solutions. Here we rethink quantized network parameterization in HW-aware fashion, towards a unified analysis of all quantization DoF, permitting for the first time their joint end-to-end finetuning. Our single-step simple and extendable method, dubbed quantization-aware finetuning (QFT), achieves 4-bit weight quantization results on-par with SoTA within PTQ constraints of speed and resource.
Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 44 of its 64 references that are in the map. First posted on arXiv 5 Dec 2022.
Builds on
44 of 64 references.
References · 44
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2015Distilling the Knowledge in a Neural Networkcited by 10,076
- 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
- 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
- 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
- 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,827
- 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,031
- 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,548
- 2020Designing Network Design Spacescited by 954
- 2018MnasNet: Platform-Aware Neural Architecture Search for Mobilecited by 937
- 2021A Survey of Quantization Methods for Efficient Neural Network Inferencecited by 798
- 2019Learned Step Size Quantizationcited by 604
32 more
- 2020Up or Down? Adaptive Rounding for Post-Training Quantizationcited by 600
- 2018Quantizing deep convolutional networks for efficient inference: A whitepapercited by 507
- 2021BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstructioncited by 464
- 2018PACT: Parameterized Clipping Activation for Quantized Neural Networkscited by 456
- 2021A White Paper on Neural Network Quantizationcited by 453
- 2016Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activationscited by 413
- 2019Data-Free Quantization Through Weight Equalization and Bias Correctioncited by 328
- 2022Optimal Brain Compression: A Framework for Accurate Post-Training Quantization and Pruningcited by 317
- 2018Model compression via distillation and quantizationcited by 287
- 2019Differentiable Soft Quantization: Bridging Full-Precision and Low-Bit Neural Networkscited by 234
- 2019Low-bit Quantization of Neural Networks for Efficient Inferencecited by 185
- 2020Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluationcited by 183
- 2019Improving Neural Network Quantization without Retraining using Outlier Channel Splittingcited by 150
- 2020Training with Quantization Noise for Extreme Model Compressioncited by 145
- 2018Scalable Methods for 8-bit Training of Neural Networkscited by 139
- 2017Apprentice: Using Knowledge Distillation Techniques To Improve Low-Precision Network Accuracycited by 100
- 2020Improving Post Training Neural Quantization: Layer-wise Calibration and Integer Programmingcited by 89
- 2019Trained Quantization Thresholds for Accurate and Efficient Fixed-Point Inference of Deep Neural Networkscited by 87
- 2017Towards Effective Low-bitwidth Convolutional Neural Networkscited by 76
- 2019Same, Same But Different - Recovering Neural Network Quantization Error Through Weight Factorizationcited by 45
- 2018Post-training 4-bit quantization of convolution networks for rapid-deploymentcited by 38
- 2021MQBench: Towards Reproducible and Deployable Model Quantization Benchmarkcited by 38
- 2018Discovering Low-Precision Networks Close to Full-Precision Networks for Efficient Embedded Inferencecited by 33
- 2019Fighting Quantization Bias With Biascited by 30
- 2022Optimal Clipping and Magnitude-aware Differentiation for Improved Quantization-aware Trainingcited by 26
- 2021Post-training deep neural network pruning via layer-wise calibrationcited by 23
- 2020Neural Network Compression Framework for fast model inferencecited by 18
- 2021Post-Training Sparsity-Aware Quantizationcited by 16
- 2019Learning low-precision neural networks without Straight-Through Estimator(STE)cited by 15
- 2020Joint Pruning & Quantization for Extremely Sparse Neural Networkscited by 8
- 2021Training Deep Neural Networks with Joint Quantization and Pruning of Weights and Activationscited by 6
- 2020Exploring Neural Networks Quantization via Layer-Wise Quantization Analysiscited by 3
Built on by
Citing papers per year: 2024: 2; 2025: 0; 2026: 1.
3 papers.