Research map / Architectures and efficiency / Quantization
Adaptive Quantization for Deep Neural Network
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Allocate quantization bit-width adaptively across layers using layer-specific accuracy sensitivity | Pre-trained ImageNet models · ImageNet validation — AlexNet and VGG-16 | model size at the same accuracy degradation: 30-40% smaller model size | Helped |
| Allocate quantization bit-width adaptively across layers using layer-specific accuracy sensitivity | Pre-trained ImageNet models · ImageNet validation — GoogLeNet and ResNet-50 | model size at the same accuracy degradation: 15-20% smaller model size | Helped |
| Use adaptive layer-wise bit-widths rather than SQNR-based or equal-bit-width quantization | Pre-trained ImageNet models · ImageNet validation — AlexNet, VGG-16, GoogLeNet, and ResNet-50 | compression rate: 20-40% higher compression rate | Helped |
| Use SQNR-based rather than equal-bit-width quantization | ResNet-50 · ImageNet validation | model size at the same accuracy degradation | No clear effect |
Closest papers by meaning
- Adaptive Quantization of Neural Networks (ICLR 2018 · similarity 0.93)
- Deep Neural Network Quantization via Layer-Wise Optimization Using Limited Training Data (AAAI 2019 · similarity 0.92)
- Explicit Loss-Error-Aware Quantization for Low-Bit Deep Neural Networks (CVPR 2018 · similarity 0.91)
- Adaptive Loss-Aware Quantization for Multi-Bit Networks (CVPR 2020 · similarity 0.89)
- AdaBits: Neural Network Quantization With Adaptive Bit-Widths (CVPR 2020 · similarity 0.89)
- Effective Interplay between Sparsity and Quantization: From Theory to Practice (ICLR 2025 · similarity 0.88)
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.