Research map / Architectures and efficiency / CNN design
Convolutional Networks with Adaptive Inference Graphs
Andreas Veit, Serge Belongie
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 |
|---|---|---|---|
| replace ResNet 110 with ConvNet-AIG 110 (gated, executes subset of layers) | ResNet 110 · CIFAR-10 | test error (%): 6.61 -> 5.76 | Improved |
| execute all layers with outputs scaled by expected execution rate (ConvNet-AIG 110*) (ablation) | ConvNet-AIG 110 · CIFAR-10 | test error (%): 5.76 -> 5.14 | Improved |
| replace ResNet 50 (ours) with ConvNet-AIG 50 at target rate 0.7 | ResNet 50 · ImageNet val | top-1 error (%): 23.87 -> 23.82 | Improved |
| replace ResNet 101 (ours) with ConvNet-AIG 101 at target rate 0.5 | ResNet 101 · ImageNet val | top-1 error (%): 22.63 -> 22.63 | No clear effect |
| lower target rate for ConvNet-AIG 50 from 0.7 to 0.4 (ablation) | ConvNet-AIG 50 · ImageNet val | top-1 error (%): 23.82 -> 24.75 | Hurt |
| lower target rate for ConvNet-AIG 50 from 0.7 to 0.4 (ablation) | ConvNet-AIG 50 · ImageNet val | FLOPs (x10^9): 3.06 -> 2.56 | Improved |
| apply stochastic depth to ResNet 50 | ResNet 50 · ImageNet val | top-1 error (%): 23.87 -> 27.75 | Hurt |
| apply stochastic depth to ResNet 101 | ResNet 101 · ImageNet val | top-1 error (%): 22.63 -> 22.80 | Improved |
| use ConvNet-AIG gating instead of plain ResNet under FGSM adversarial attack | ResNet · ImageNet val (FGSM) | top-1 accuracy | Improved |
| always execute early layers in ConvNet-AIG 101 (ablation) | ConvNet-AIG 101 · ImageNet val | design choice | No clear effect |
Closest papers by meaning
- Adaptive Neural Networks for Efficient Inference (ICML 2017 · similarity 0.89)
- EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks (ICML 2019 · similarity 0.87)
- AdaGCN: Adaboosting Graph Convolutional Networks into Deep Models (ICLR 2021 · similarity 0.87)
- Feedback-Prop: Convolutional Neural Network Inference Under Partial Evidence (CVPR 2018 · similarity 0.86)
- On the Stability of Graph Convolutional Neural Networks: A Probabilistic Perspective (NeurIPS 2025 · similarity 0.86)
- Convolutional Normalization: Improving Deep Convolutional Network Robustness and Training (NeurIPS 2021 · similarity 0.86)
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.
