Research map / Architectures and efficiency / RNNs and normalization
Arboreal Neural Network
Wubin Yan, Wei Ma, Shixiang Wan, Dongchen Li, Shaoshun Kang, Qing Yang, Dongliang Xu
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 |
|---|---|---|---|
| fine-tune only split thresholds and leaf values (f+v) instead of leaving the decompiled tree untrained (ablation) | ArbNN · TabCredit M2+@MOB3 | KS (%): 53.92 -> 54.13 | Improved |
| unfreeze all parameters instead of only f+v (ablation) | ArbNN · TabCredit M2+@MOB3 | KS (%): 54.13 -> 53.69 | Hurt |
| restrict fine-tuning to f+v and keep tree-structure matrix P frozen instead of fine-tuning P (ablation) | ArbNN · TabCredit M2+@MOB3 | KS (%): 53.94 -> 54.13 | Improved |
| initialize f and v from the decompiled XGBoost tree instead of randomly (ablation) | ArbNN · TabCredit M2+@MOB3 | KS (%): 53.95 -> 54.13 | Improved |
| random initialization of f and v still beats the XGBoost baseline (ablation) | ArbNN (random init) · TabCredit M2+@MOB3 | KS (%): 53.92 -> 53.95 | Improved |
| replace XGBoost with ArbNN | XGBoost · TabCredit OOT test | ECE: 0.0138 -> 0.0011 | Improved |
| replace XGBoost with ArbNN | XGBoost · TabCredit OOT test | Brier: 0.0246 -> 0.0241 | Improved |
| replace XGBoost with ArbNN | XGBoost · TabCredit OOT test | AUC-ROC: 0.8188 -> 0.8194 | Improved |
| replace XGBoost with ArbNN | XGBoost · TabCredit vintage M2+@MOB12 | KS (%): 39.13 -> 39.32 | Improved |
| temperature sweep tau1,tau2 in {5,10,50,100,200}; choose tau1=100, tau2=10 (ablation) | ArbNN · tabular benchmarks | accuracy/smoothness trade-off | Improved |
Closest papers by meaning
- MetaSymNet: A Tree-like Symbol Network with Adaptive Architecture and Activation Functions (AAAI 2025 · similarity 0.87)
- Neural Reasoning Networks: Efficient Interpretable Neural Networks with Automatic Textual Explanations (AAAI 2025 · similarity 0.85)
- SMILE: Bridging Continuous Optimization and Discrete Symbolic Recovery (NeurIPS 2026 · similarity 0.85)
- SHAP Meets Tensor Networks: Provably Tractable Explanations with Parallelism (NeurIPS 2025 · similarity 0.84)
- Mastering Symbolic Operations: Augmenting Language Models with Compiled Neural Networks (ICLR 2024 · similarity 0.84)
- Emergent Symbolic Mechanisms Support Abstract Reasoning in Large Language Models (ICML 2025 · similarity 0.84)
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.
