mlsanityData to 14 Oct 2026

AutoQNN: An End-to-End Framework for Automatically Quantizing Neural Networks

arXiv 2304.03782 ↗First posted 7 Apr 2023cs.LG cs.AI cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Exploring the expected quantizing scheme with suitable mixed-precision policy is the key point to compress deep neural networks (DNNs) in high efficiency and accuracy. This exploration implies heavy workloads for domain experts, and an automatic compression method is needed. However, the huge search space of the automatic method introduces plenty of computing budgets that make the automatic process challenging to be applied in real scenarios. In this paper, we propose an end-to-end framework named AutoQNN, for automatically quantizing different layers utilizing different schemes and bitwidths without any human labor. AutoQNN can seek desirable quantizing schemes and mixed-precision policies for mainstream DNN models efficiently by involving three techniques: quantizing scheme search (QSS), quantizing precision learning (QPL), and quantized architecture generation (QAG). QSS introduces five quantizing schemes and defines three new schemes as a candidate set for scheme search, and then uses the differentiable neural architecture search (DNAS) algorithm to seek the layer- or model-desired scheme from the set. QPL is the first method to learn mixed-precision policies by reparameterizing the bitwidths of quantizing schemes, to the best of our knowledge. QPL optimizes both classification loss and precision loss of DNNs efficiently and obtains the relatively optimal mixed-precision model within limited model size and memory footprint. QAG is designed to convert arbitrary architectures into corresponding quantized ones without manual intervention, to facilitate end-to-end neural network quantization. We have implemented AutoQNN and integrated it into Keras. Extensive experiments demonstrate that AutoQNN can consistently outperform state-of-the-art quantization.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 43 of its 67 references that are in the map. First posted on arXiv 7 Apr 2023.

Builds on

43 of 67 references.

References · 43

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  4. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,475
  5. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,795
  6. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,845
  7. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,739
  8. 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,593
  9. 2016The Concrete Distribution: A Continuous Relaxation of Discrete Random Variablescited by 1,148
  10. 2016XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networkscited by 849
  11. 2015BinaryConnect: Training Deep Neural Networks with binary weights during propagationscited by 592
  12. 2016DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradientscited by 569
31 more
  1. 2018PACT: Parameterized Clipping Activation for Quantized Neural Networkscited by 467
  2. 2018HAQ: Hardware-Aware Automated Quantization with Mixed Precisioncited by 443
  3. 2016Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activationscited by 426
  4. 2019HAWQ: Hessian AWare Quantization of Neural Networks with Mixed-Precisioncited by 402
  5. 2014Deeply-Supervised Netscited by 367
  6. 2018LQ-Nets: Learned Quantization for Highly Accurate and Compact Deep Neural Networkscited by 274
  7. 2019HAWQ-V2: Hessian Aware trace-Weighted Quantization of Neural Networkscited by 261
  8. 2019Differentiable Soft Quantization: Bridging Full-Precision and Low-Bit Neural Networkscited by 238
  9. 2017Incremental Network Quantization: Towards Lossless CNNs with Low-Precision Weightscited by 217
  10. 2016Trained Ternary Quantizationcited by 208
  11. 2016Ternary Weight Networkscited by 204
  12. 2018Learning to Quantize Deep Networks by Optimizing Quantization Intervals with Task Losscited by 160
  13. 2017Towards Accurate Binary Convolutional Neural Networkcited by 149
  14. 2015Fixed Point Quantization of Deep Convolutional Networkscited by 143
  15. 2018Mixed Precision Quantization of ConvNets via Differentiable Neural Architecture Searchcited by 125
  16. 2016Convolutional Neural Networks using Logarithmic Data Representationcited by 99
  17. 2019AutoQ: Automated Kernel-Wise Neural Network Quantizationcited by 63
  18. 2017Extremely Low Bit Neural Network: Squeeze the Last Bit Out with ADMMcited by 61
  19. 2015Neural Networks with Few Multiplicationscited by 48
  20. 2021BSQ: Exploring Bit-Level Sparsity for Mixed-Precision Neural Network Quantizationcited by 39
  21. 2016Ternary Neural Networks for Resource-Efficient AI Applicationscited by 33
  22. 2017Performance Guaranteed Network Acceleration via High-Order Residual Quantizationcited by 26
  23. 2017Balanced Quantization: An Effective and Efficient Approach to Quantized Neural Networkscited by 21
  24. 2018Blended Coarse Gradient Descent for Full Quantization of Deep Neural Networkscited by 21
  25. 2020Binarizing MobileNet via Evolution-based Searchingcited by 21
  26. 2016Hardware-oriented Approximation of Convolutional Neural Networkscited by 20
  27. 2020VecQ: Minimal Loss DNN Model Compression With Vectorized Weight Quantizationcited by 15
  28. 2016Effective Quantization Methods for Recurrent Neural Networkscited by 14
  29. 2016ImageNet pre-trained models with batch normalizationcited by 14
  30. 2016Ristretto: Hardware-Oriented Approximation of Convolutional Neural Networkscited by 6
  31. 2017Low Precision RNNs: Quantizing RNNs Without Losing Accuracycited by 5

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2022A Comprehensive Survey on Model Quantization for Deep Neural Networks in Image Classification
  2. 2024Differentiable Search for Finding Optimal Quantization Strategy
  3. 2024Magic for the Age of Quantized DNNs
  4. 2024AutoMixQ: Self-Adjusting Quantization for High Performance Memory-Efficient Fine-Tuning
  5. 2023A Practical Mixed Precision Algorithm for Post-Training Quantization
  6. 2022QFT: Post-training quantization via fast joint finetuning of all degrees of freedom
  7. 2021Towards Mixed-Precision Quantization of Neural Networks via Constrained Optimization
  8. 2025Towards Training-free Automatic Proxy Discovery via Large Language Models for Mixed Precision Quantization