Research map / Detection and segmentation / CNN features
Interpreting CNNs via Decision Trees
Quanshi Zhang, Yu Yang, Haotian Ma, Ying Nian Wu
Check 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 leaf-level rather than 2nd-layer decision modes (ablation) | VGG-16 decision tree · ILSVRC Animal-Part | fitness of contribution distributions: 0.23 -> 1.00 | Improved |
| Use leaf-level rather than 2nd-layer decision modes (ablation) | VGG-16 decision tree · VOC Part | fitness of contribution distributions: 0.22 -> 1.00 | Improved |
| Use leaf-level rather than 2nd-layer decision modes (ablation) | VGG-16 decision tree · CUB200-2011 | fitness of contribution distributions: 0.21 -> 1.00 | Improved |
| Use leaf-level rather than 2nd-layer decision modes (ablation) | VGG-16 decision tree · ILSVRC Animal-Part | classification accuracy: 94.4 -> 87.8 | Hurt |
| Use leaf-level rather than 2nd-layer decision modes (ablation) | VGG-16 decision tree · VOC Part | classification accuracy: 94.2 -> 88.2 | Hurt |
| Use leaf-level rather than 2nd-layer decision modes (ablation) | VGG-16 decision tree · CUB200-2011 | classification accuracy: 91.5 -> 85.3 | Hurt |
| Use leaf-level rather than 2nd-layer decision modes (ablation) | VGG-16 decision tree · CUB200-2011 | prediction error: 0.075 -> 0.00 | Improved |
Closest papers by meaning
- Training Decision Trees as Replacement for Convolution Layers (AAAI 2020 · similarity 0.89)
- End-to-End Learning of Decision Trees and Forests (IJCV 2019 · similarity 0.87)
- Learning Decision Trees Recurrently Through Communication (CVPR 2021 · similarity 0.87)
- ViTree: Single-Path Neural Tree for Step-Wise Interpretable Fine-Grained Visual Categorization (AAAI 2024 · similarity 0.86)
- Finding Representative Interpretations on Convolutional Neural Networks (ICCV 2021 · similarity 0.86)
- Comparing the Decision-Making Mechanisms by Transformers and CNNs via Explanation Methods (CVPR 2024 · 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.
