Research map / Label-efficient and robust learning / Recognition and long-tail
Neural Bootstrapper
Minsuk Shin, Hyungjoo Cho, Hyun-seok Min, Sungbin Lim
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 NeuBoots instead of amortized bootstrapping with B=5 | Rotated MNIST classification · Rotated MNIST | test error: 16.82 -> 14.98±0.31 | Improved |
| Increase bootstrap samples from B=1 to B=25 (ablation) | NeuBoots · Rotated MNIST | test error: 17.94±0.74 -> 14.45±0.31 | Improved |
| Add calibration to NeuBoots (ablation) | CIFAR-10 to SVHN OOD detection · CIFAR-10 to SVHN OOD | AUROC: 97.26 -> 99.14 | Improved |
| Use NeuBoots instead of DeepEnsemble | PASCAL VOC 2012 semantic segmentation, ResNet-50 · PASCAL VOC 2012 | mIoU (%): 90.09±0.61 -> 90.14±2.17 | No clear effect |
| Use NeuBoots instead of DeepEnsemble_CE | NIH3T3 semantic segmentation, U-ResNet · NIH3T3 | mIoU (%): 65.71±1.69 -> 67.78±1.01 | Improved |
| Use NeuBoots instead of DeepEnsemble_CE | NIH3T3 semantic segmentation, SCNAS · NIH3T3 | mIoU (%): 68.66±2.58 -> 70.80±1.58 | Improved |
| Use NeuBoots instead of DeepEnsemble | CIFAR-100 calibration, ResNet-110 · CIFAR-100 | ECE (%): 4.74±0.17 -> 8.13±0.28 | Hurt |
| Use NeuBoots instead of MCDrop | CIFAR-100 prediction · CIFAR-100 | relative prediction time: 5.0 -> 1.2 | Improved |
| Use NeuBoots instead of DeepEnsemble | CIFAR-10 calibration, DenseNet-100 · CIFAR-10 | ECE (%): 0.54 -> 2.87 | Hurt |
| Use NeuBoots instead of MCDrop | CIFAR-10 test-time bootstrap outputs, ResNet-34 · CIFAR-10 test set | prediction time for 100 outputs (s): 112s -> 1.9s | Improved |
| Use NeuBoots instead of DeepEnsemble | ResNet-110 training · CIFAR-100 | training time (sec): 145.5 -> 29.2 | Improved |
Closest papers by meaning
- Bootstrapping neural processes (NeurIPS 2020 · similarity 0.87)
- Uncertainty Quantification with the Empirical Neural Tangent Kernel (NeurIPS 2025 · similarity 0.83)
- Orthogonal Bootstrap: Efficient Simulation of Input Uncertainty (ICML 2024 · similarity 0.83)
- L2B: Learning to Bootstrap Robust Models for Combating Label Noise (CVPR 2024 · similarity 0.83)
- Make Me a BNN: A Simple Strategy for Estimating Bayesian Uncertainty from Pre-trained Models (CVPR 2024 · similarity 0.82)
- Spend Wisely: Maximizing Post-Training Gains in Iterative Synthetic Data Bootstrapping (NeurIPS 2025 · similarity 0.82)
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.
