Research map / Architectures and efficiency / Quantization
SDQ: Stochastic Differentiable Quantization with Mixed Precision
Xijie Huang, Zhiqiang Shen, Shichao Li, Zechun Liu, Hu Xianghong, Jeffry Wicaksana, Eric Xing, Kwang-Ting Cheng
Read the paperCheck your idea against 137,217 papers
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 |
|---|---|---|---|
| use stochastic differentiable bitwidth selection instead of the strongest listed fixed-precision or MPQ baselines | ResNet20 · CIFAR-10 | Top-1 accuracy (%): 91.9 -> 92.1 | Improved |
| use SDQ mixed precision instead of DDQ | ResNet18 · ImageNet-1K | Top-1 accuracy (%): 71.2 -> 71.7 | Improved |
| use SDQ mixed precision instead of DDQ | MobileNetV2 · ImageNet-1K | Top-1 accuracy (%): 71.8 -> 72.0 | Improved |
| use SDQ strategy generation rather than HMQ strategy generation (ablation) | MobileNetV2 · ImageNet-1K | Top-1 accuracy (%): 71.7 -> 72.0 | Improved |
| use SDQ strategy generation rather than FracBits strategy generation (ablation) | MobileNetV2 · ImageNet-1K | Top-1 accuracy (%): 72.0 -> 72.0 | No clear effect |
| add EBR with lambda 0.1 (ablation) | mixed-precision ResNet18 · ImageNet-1K | Top-1 accuracy (%): 67.6 -> 69.1 | Improved |
| replace EBR with Weight Norm (ablation) | mixed-precision ResNet18 · ImageNet-1K | Top-1 accuracy (%): 69.1 -> 66.6 | Hurt |
| vary EBR weight from 0.1 to 0.01 (ablation) | mixed-precision ResNet18 · ImageNet-1K | Top-1 accuracy (%): 69.1 -> 68.6 | Hurt |
| vary EBR weight from 0.1 to 1 (ablation) | mixed-precision ResNet18 · ImageNet-1K | Top-1 accuracy (%): 69.1 -> 68.9 | Hurt |
| add knowledge distillation from a ResNet101 teacher (ablation) | mixed-precision ResNet18 · ImageNet-1K | Top-1 accuracy (%): 70.5 -> 71.7 | Improved |
| use a stronger ResNet101 teacher rather than ResNet34 (ablation) | mixed-precision ResNet18 · ImageNet-1K | Top-1 accuracy (%): 70.7 -> 71.7 | Improved |
| use block-wise rather than layer-wise quantization (ablation) | ResNet18 · ImageNet-1K | Top-1 accuracy (%): 71.7 -> 71.2 | Hurt |
Closest papers by meaning
- Retraining-Free Model Quantization via One-Shot Weight-Coupling Learning (CVPR 2024 · similarity 0.89)
- MPQ-DM: Mixed Precision Quantization for Extremely Low Bit Diffusion Models (AAAI 2025 · similarity 0.89)
- ParetoQ: Improving Scaling Laws in Extremely Low-bit LLM Quantization (NeurIPS 2025 · similarity 0.88)
- PTMQ: Post-training Multi-Bit Quantization of Neural Networks (AAAI 2024 · similarity 0.88)
- GSQ: Highly-Accurate Low-Precision Scalar Quantization for LLMs via Gumbel-Softmax Sampling (NeurIPS 2026 · similarity 0.88)
- Learning from Loss Landscape: Generalizable Mixed-Precision Quantization via Adaptive Sharpness-Aware Gradient Aligning (ICML 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.
