Research map / Probabilistic and scientific modeling / Bayesian optimization
Demystifying Black-box Models with Symbolic Metamodels
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 Meijer G-function symbolic metamodel (closed-form) instead of genetic-programming symbolic regression | function fitting · Bessel J0(10 sqrt(x)) | R2: 0.116 -> 0.999 | Helped |
| use closed-form metamodel instead of symbolic regression | function fitting · exp(-3x) | R2: 0.970 -> 0.999 | Helped |
| use closed-form metamodel instead of symbolic regression | function fitting · sin(x) | R2: 0.998 -> 0.999 | No clear effect |
| restrict metamodel class from closed-form to polynomial (ablation) | symbolic metamodel · Bessel J0(10 sqrt(x)) | R2: 0.999 -> -4.75 | Hurt |
| restrict metamodel class from closed-form to polynomial (ablation) | symbolic metamodel · exp(-3x) | R2: 0.999 -> 0.995 | Hurt |
| use polynomial symbolic metamodel instead of symbolic regression | function fitting · exp(-3x), x/(x+1)^2, sin(x), J0 | R2: 0.995 vs 0.970; -4.75 vs 0.116 | Mixed |
| derive instancewise feature importance from metamodel Taylor expansion vs DeepLIFT/SHAP/LIME/L2X | symbolic metamodel · XOR, nonlinear additive, switch | median feature rank | Mixed |
| use symbolic metamodel instead of symbolic regression for feature ranking | feature importance · synthetic datasets | median feature rank | Helped |
Closest papers by meaning
- SYMBOL: Generating Flexible Black-Box Optimizers through Symbolic Equation Learning (ICLR 2024 · similarity 0.86)
- MetaSymNet: A Tree-like Symbol Network with Adaptive Architecture and Activation Functions (AAAI 2025 · similarity 0.86)
- Symbolic regression via MDLformer-guided search: from minimizing prediction error to minimizing description length (ICLR 2025 · similarity 0.85)
- Ab Initio Nonparametric Variable Selection for Scalable Symbolic Regression with Large $p$ (ICML 2025 · similarity 0.85)
- SMILE: Bridging Continuous Optimization and Discrete Symbolic Recovery (NeurIPS 2026 · similarity 0.84)
- Local vs. Global Interpretability: A Computational Complexity Perspective (ICML 2024 · 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.