# HAWQV3: Dyadic Neural Network Quantization

arXiv: https://arxiv.org/abs/2011.10680 · HTML: https://mlsanity.com/paper/2011.10680/

## Key facts

- Cited by 60 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Nov 2020.

## Built on by (most cited first, 30 of 60)

- [A Survey of Quantization Methods for Efficient Neural Network Inference](https://mlsanity.com/paper/2103.13630/) (2021)
- [SqueezeLLM: Dense-and-Sparse Quantization](https://mlsanity.com/paper/2306.07629/) (2023)
- [I-BERT: Integer-only BERT Quantization](https://mlsanity.com/paper/2101.01321/) (2021)
- [Full Stack Optimization of Transformer Inference: a Survey](https://mlsanity.com/paper/2302.14017/) (2023)
- [NoisyQuant: Noisy Bias-Enhanced Post-Training Activation Quantization for Vision Transformers](https://mlsanity.com/paper/2211.16056/) (2022)
- [Differentiable Model Compression via Pseudo Quantization Noise](https://mlsanity.com/paper/2104.09987/) (2021)
- [A Survey of Techniques for Optimizing Transformer Inference](https://mlsanity.com/paper/2307.07982/) (2023)
- [Applications and Techniques for Fast Machine Learning in Science](https://mlsanity.com/paper/2110.13041/) (2021)
- [SDQ: Stochastic Differentiable Quantization with Mixed Precision](https://mlsanity.com/paper/2206.04459/) (2022)
- [Sharpness-aware Quantization for Deep Neural Networks](https://mlsanity.com/paper/2111.12273/) (2021)
- [EMQ: Evolving Training-free Proxies for Automated Mixed Precision Quantization](https://mlsanity.com/paper/2307.10554/) (2023)
- [Integer-only Zero-shot Quantization for Efficient Speech Recognition](https://mlsanity.com/paper/2103.16827/) (2021)
- [HAO: Hardware-aware neural Architecture Optimization for Efficient Inference](https://mlsanity.com/paper/2104.12766/) (2021)
- [Patch-wise Mixed-Precision Quantization of Vision Transformer](https://mlsanity.com/paper/2305.06559/) (2023)
- [Neural Network Quantization for Efficient Inference: A Survey](https://mlsanity.com/paper/2112.06126/) (2021)
- [QT-DoG: Quantization-aware Training for Domain Generalization](https://mlsanity.com/paper/2410.06020/) (2024)
- [BMPQ: Bit-Gradient Sensitivity Driven Mixed-Precision Quantization of DNNs from Scratch](https://mlsanity.com/paper/2112.13843/) (2021)
- [Flexible and Fully Quantized Ultra-Lightweight TinyissimoYOLO for Ultra-Low-Power Edge Systems](https://mlsanity.com/paper/2307.05999/) (2023)
- [Low-bit Model Quantization for Deep Neural Networks: A Survey](https://mlsanity.com/paper/2505.05530/) (2025)
- [Confounding Tradeoffs for Neural Network Quantization](https://mlsanity.com/paper/2102.06366/) (2021)
- [Finding the Task-Optimal Low-Bit Sub-Distribution in Deep Neural Networks](https://mlsanity.com/paper/2112.15139/) (2021)
- [Bitwidth-Adaptive Quantization-Aware Neural Network Training: A Meta-Learning Approach](https://mlsanity.com/paper/2207.10188/) (2022)
- [DeepGEMM: Accelerated Ultra Low-Precision Inference on CPU Architectures using Lookup Tables](https://mlsanity.com/paper/2304.09049/) (2023)
- [MetaMix: Meta-state Precision Searcher for Mixed-precision Activation Quantization](https://mlsanity.com/paper/2311.06798/) (2023)
- [TransAxx: Efficient Transformers with Approximate Computing](https://mlsanity.com/paper/2402.07545/) (2024)
- [eXmY: A Data Type and Technique for Arbitrary Bit Precision Quantization](https://mlsanity.com/paper/2405.13938/) (2024)
- [Neuroevolution-Enhanced Multi-Objective Optimization for Mixed-Precision Quantization](https://mlsanity.com/paper/2106.07611/) (2021)
- [FLIQS: One-Shot Mixed-Precision Floating-Point and Integer Quantization Search](https://mlsanity.com/paper/2308.03290/) (2023)
- [Low-bit Quantization for Deep Graph Neural Networks with Smoothness-aware Message Propagation](https://mlsanity.com/paper/2308.14949/) (2023)
- [M4BRAM: Mixed-Precision Matrix-Matrix Multiplication in FPGA Block RAMs](https://mlsanity.com/paper/2311.02758/) (2023)

## Abstract (arXiv)

Current low-precision quantization algorithms often have the hidden cost of conversion back and forth from floating point to quantized integer values. This hidden cost limits the latency improvement realized by quantizing Neural Networks. To address this, we present HAWQV3, a novel mixed-precision integer-only quantization framework. The contributions of HAWQV3 are the following: (i) An integer-only inference where the entire computational graph is performed only with integer multiplication, addition, and bit shifting, without any floating point operations or even integer division; (ii) A novel hardware-aware mixed-precision quantization method where the bit-precision is calculated by solving an integer linear programming problem that balances the trade-off between model perturbation and other constraints, e.g., memory footprint and latency; (iii) Direct hardware deployment and open source contribution for 4-bit uniform/mixed-precision quantization in TVM, achieving an average speed up of 1.45× for uniform 4-bit, as compared to uniform 8-bit for ResNet50 on T4 GPUs; and (iv) extensive evaluation of the proposed methods on ResNet18/50 and InceptionV3, for various model compression levels with/without mixed precision. For ResNet50, our INT8 quantization achieves an accuracy of 77.58%, which is 2.68% higher than prior integer-only work, and our mixed-precision INT4/8 quantization can reduce INT8 latency by 23% and still achieve 76.73% accuracy. Our framework and the TVM implementation have been open sourced.