Research map / Adversarial robustness and security / Certified and empirical robustness
Double-Win Quant: Aggressively Winning Robustness of Quantized Deep Neural Networks via Random Precision Training and Inference
Yonggan Fu, Qixuan Yu, Meng Li, Vikas Chandra, Yingyan Lin
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 |
|---|---|---|---|
| add RPI (random precision inference) to adversarially trained model (ablation) | FGSM, FGSM-RS, PGD-7 on PreActResNet18/WideResNet32 · CIFAR-10 PGD-20 | robust accuracy: e.g. PGD-7 PreActResNet18 51.17 -> 57.09 | Improved |
| add RPT (random precision training with switchable BN) on top of RPI (ablation) | PGD-7 / FGSM / FGSM-RS DWQ · CIFAR-10 PGD-20 | robust accuracy: PGD-7 PreActResNet18 57.09 -> 65.15 | Improved |
| add RPI+RPT on top of PGD-7 training | PreActResNet18 CIFAR-100 · CIFAR-100 PGD-20 | robust accuracy: 27.97 -> 41.74 | Improved |
| add RPI+RPT on top of FGSM-RS on ImageNet ResNet-50 while reducing BitOPs | ResNet-50 ImageNet · ImageNet PGD-10 | robust accuracy: 30.28 -> 37.93 | Improved |
| add RPI+RPT on top of PGD-7 on MobileNetV2 (compact model) | MobileNetV2 CIFAR-10 · CIFAR-10 PGD-20 | robust accuracy vs natural accuracy: robust 50.03 -> 57.72 but natural 80.40 -> 73.97 | Mixed |
| add RPI+RPT on top of PGD-7 under larger epsilon | PreActResNet18/WideResNet32/MobileNetV2 CIFAR-10 · CIFAR-10 eps=12 and 16 PGD-20 | robust accuracy: gains of 11.66%-18.89% (eps=12) and 15.68%-23.26% (eps=16) | Improved |
| add RPT with switchable BN vs CPT (cyclic precision training) on top of PGD-7 (ablation) | PreActResNet18 CIFAR-10 · CIFAR-10 PGD-20 | robust accuracy: CPT-32 66.40 vs RPT 65.15 on PGD | No clear effect |
| replace RPT with CPT on FGSM and FGSM-RS training (ablation) | PreActResNet18 CIFAR-10 · CIFAR-10 PGD-20 | robust accuracy: FGSM CPT-16 35.45 vs RPT 64.08 | Hurt |
| RPI with different training precisions (fixed-precision adversarial training) on PGD-7 networks (ablation) | three PGD-7 networks · CIFAR-10 PGD-20 | natural and robust accuracy stability: within 1.12% | No clear effect |
| RPI with smaller precision ranges and precisions close to training precision (no RPT) (ablation) | PGD-7 + RPI · CIFAR-10 PGD-20 | robust accuracy: slightly favors smaller precision ranges | Mixed |
| add RPI+RPT on top of PGD-7 against stronger attacks (CW-Inf, Bandits, Auto-Attack) | PreActResNet18/WideResNet32 CIFAR-10/100 · CW-Inf, Bandits, Auto-Attack | robust accuracy: +5.01% to +25.71% across attacks | Improved |
| add RPI+RPT on top of PGD-7 comparison to Lin et al. 2019 reported results | PGD-7 CIFAR-10 · CIFAR-10 PGD-20 eps=8 and 16 | robust accuracy: 14.6% and 22.5% higher | Improved |
Closest papers by meaning
- Defensive Quantization: When Efficiency Meets Robustness (ICLR 2019 · similarity 0.91)
- Improved Gradient-Based Adversarial Attacks for Quantized Networks (AAAI 2022 · similarity 0.88)
- Quantization-Aware Interval Bound Propagation for Training Certifiably Robust Quantized Neural Networks (AAAI 2023 · similarity 0.88)
- Qu-ANTI-zation: Exploiting Quantization Artifacts for Achieving Adversarial Outcomes (NeurIPS 2021 · similarity 0.87)
- Theoretical Analysis of Robust Overfitting for Wide DNNs: An NTK Approach (ICLR 2024 · similarity 0.86)
- Effective Interplay between Sparsity and Quantization: From Theory to Practice (ICLR 2025 · similarity 0.85)
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.
