Research map / Architectures and efficiency / Quantization
QPP: Real-Time Quantization Parameter Prediction for Deep Neural Networks
Vladimir Kryzhanovskiy, Gleb Balitskiy, Nikolay Kozyrskiy, Aleksandr Zuruev
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Predict per-sample QPs with QPP instead of using averaged static QPs | D+S quantization · ImageNet, CityScapes, COFW, WFLW, 300W, Vimeo, Set5, Set14 | task performance (Top-1 accuracy, mIoU, NME, or PSNR): 69.08 -> 69.14; 72.60 -> 72.67; 69.99 -> 70.0; 3.52 -> 3.55; 4.68 -> 4.69; 3.92 -> 3.93; 34.0 -> 34.01; 30.70 -> 30.69; | Mixed |
| Predict QPs with QPP rather than averaged static QPs | D+S quantization · All evaluated task datasets | FR(1): 3.4 / 16.1 -> 0.78 / 7.7; 4.5 / 23.9 -> 0.5 / 7.1; 0.7 / 13.9 -> 0.06 / 2.2; 3.5 / 64.2 -> 0.2 / 56.7; 8.5 / 93.4 -> 1.4 | Improved |
| Use D+S quantization instead of standard dynamic quantization | Post-training 4-bit activation quantization · ImageNet (ResNet34) | Top-1 accuracy (%): 70.34 -> 72.72 | Improved |
| Replace dynamic quantile calculation with QPP | D+S quantization inference · Kirin 980 convolution benchmark | inference time: 23% overhead; decreased by 11% | Improved |
| Use D+S QPP rather than full-precision inference | Convolution inference on Kirin 980 · Kirin 980 convolution benchmark | acceleration: x1.46 | Improved |
Closest papers by meaning
- Adaptive Quantization for Deep Neural Network (AAAI 2018 · similarity 0.88)
- Learning to Quantize Deep Networks by Optimizing Quantization Intervals With Task Loss (CVPR 2019 · similarity 0.88)
- Deep Neural Network Quantization via Layer-Wise Optimization Using Limited Training Data (AAAI 2019 · similarity 0.88)
- Scalable methods for 8-bit training of neural networks (NeurIPS 2018 · similarity 0.87)
- Efficient Quantization for Neural Networks with Binary Weights and Low Bitwidth Activations (AAAI 2019 · similarity 0.87)
- Quantization without Tears (CVPR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
