mlsanityData to 7 Oct 2026

HMQ: Hardware Friendly Mixed Precision Quantization Block for CNNs

arXiv 2007.09952 ↗First posted 20 Jul 2020cs.LG cs.CV stat.MLcited by 43

Not on the map. Why?

Abstract (arXiv)

Recent work in network quantization produced state-of-the-art results using mixed precision quantization. An imperative requirement for many efficient edge device hardware implementations is that their quantizers are uniform and with power-of-two thresholds. In this work, we introduce the Hardware Friendly Mixed Precision Quantization Block (HMQ) in order to meet this requirement. The HMQ is a mixed precision quantization block that repurposes the Gumbel-Softmax estimator into a smooth estimator of a pair of quantization parameters, namely, bit-width and threshold. HMQs use this to search over a finite space of quantization schemes. Empirically, we apply HMQs to quantize classification models trained on CIFAR10 and ImageNet. For ImageNet, we quantize four different architectures and show that, in spite of the added restrictions to our quantization scheme, we achieve competitive and, in some cases, state-of-the-art results.

Key facts. Cited by 43 papers in the map (data through 7 Oct 2026). First posted on arXiv 20 Jul 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 7; 2022: 9; 2023: 14; 2024: 6; 2025: 6; 2026: 1.

43 papers.

Sort:

  1. 2021A Survey of Quantization Methods for Efficient Neural Network Inferencecited by 798
  2. 2021Implicit Neural Representations for Image Compressioncited by 161
  3. 2022A Comprehensive Survey on Model Quantization for Deep Neural Networks in Image Classificationcited by 64
  4. 2022Mixed-Precision Neural Network Quantization via Learned Layer-wise Importancecited by 53
  5. 2022Better Teacher Better Student: Dynamic Prior Knowledge for Knowledge Distillationcited by 41
  6. 2022SDQ: Stochastic Differentiable Quantization with Mixed Precisioncited by 27
  7. 2022F8Net: Fixed-Point 8-bit Only Multiplication for Network Quantizationcited by 26
  8. 2022CADyQ: Content-Aware Dynamic Quantization for Image Super-Resolutioncited by 26
  9. 2023EMQ: Evolving Training-free Proxies for Automated Mixed Precision Quantizationcited by 22
  10. 2021Differentiable Dynamic Quantization with Mixed Precision and Adaptive Resolutioncited by 19
  11. 2024Retraining-free Model Quantization via One-Shot Weight-Coupling Learningcited by 13
  12. 2024SketchINR: A First Look into Sketches as Implicit Neural Representationscited by 13
31 more
  1. 2021Generalizable Mixed-Precision Quantization via Attribution Rank Preservationcited by 11
  2. 2023Mixed-Precision Quantization for Federated Learning on Resource-Constrained Heterogeneous Devicescited by 11
  3. 2024Channel-Wise Mixed-Precision Quantization for Large Language Modelscited by 11
  4. 2025Low-bit Model Quantization for Deep Neural Networks: A Surveycited by 11
  5. 2023EQ-Net: Elastic Quantization Neural Networkscited by 10
  6. 2021HPTQ: Hardware-Friendly Post Training Quantizationcited by 9
  7. 2023Revisiting Block-based Quantisation: What is Important for Sub-8-bit LLM Inference?cited by 8
  8. 2022Mixed-Precision Neural Networks: A Surveycited by 7
  9. 2023MetaMix: Meta-state Precision Searcher for Mixed-precision Activation Quantizationcited by 7
  10. 2021FAT: Learning Low-Bitwidth Parametric Representation via Frequency-Aware Transformationcited by 6
  11. 2021Neuroevolution-Enhanced Multi-Objective Optimization for Mixed-Precision Quantizationcited by 6
  12. 2023SEAM: Searching Transferable Mixed-Precision Quantization Policy through Large Margin Regularizationcited by 6
  13. 2023FemtoDet: An Object Detection Baseline for Energy Versus Performance Tradeoffscited by 5
  14. 2023Towards Efficient In-memory Computing Hardware for Quantized Neural Networks: State-of-the-art, Open Challenges and Perspectivescited by 5
  15. 2023EPTQ: Enhanced Post-Training Quantization via Hessian-guided Network-wise Optimizationcited by 5
  16. 2023QBitOpt: Fast and Accurate Bitwidth Reallocation during Trainingcited by 4
  17. 2023Overcoming Distribution Mismatch in Quantizing Image Super-Resolution Networkscited by 4
  18. 2025Learning from Loss Landscape: Generalizable Mixed-Precision Quantization via Adaptive Sharpness-Aware Gradient Aligningcited by 3
  19. 2024Data Generation for Hardware-Friendly Post-Training Quantizationcited by 2
  20. 2026dMX: Differentiable Mixed-Precision Assignment for Low-Precision Floating-Point Formatscited by 2
  21. 2023Learning Accurate Performance Predictors for Ultrafast Automated Model Compressioncited by 1
  22. 2023On Calibration of Modern Quantized Efficient Neural Networkscited by 1
  23. 2024CARIn: Constraint-Aware and Responsive Inference on Heterogeneous Devices for Single- and Multi-DNN Workloadscited by 1
  24. 2025MLoRQ: Bridging Low-Rank and Quantization for Transformer Compressioncited by 1
  25. 2025Where and How to Enhance: Discovering Bit-Width Contribution for Mixed Precision Quantizationcited by 1
  26. 2025FlexiQ: Adaptive Mixed-Precision Quantization for Latency/Accuracy Trade-Offs in Deep Neural Networkscited by 1
  27. 2022A Closer Look at Hardware-Friendly Weight Quantizationcited by 0
  28. 2022CSMPQ:Class Separability Based Mixed-Precision Quantizationcited by 0
  29. 2023Efficient Online Processing with Deep Neural Networkscited by 0
  30. 2024Efficient Neural Compression with Inference-time Decodingcited by 0
  31. 2025CHORD: Customizing Hybrid-precision On-device Model for Sequential Recommendation with Device-cloud Collaborationcited by 0