Research map / Learning theory and algorithms / Prediction and decision losses
Contemporary Symbolic Regression Methods and their Relative Performance
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 Operon rather than the other evaluated methods | Black-box regression methods · Black-box regression problems (122 datasets) | test-set R² | Helped |
| Use FEAT rather than LightGBM | Black-box regression · Black-box regression problems (122 datasets) | test-set R² | Helped |
| Use SBP-GP rather than XGBoost | Black-box regression · Black-box regression problems (122 datasets) | test-set R² | Helped |
| Use AIFeynman for low-noise ground-truth equation recovery | Symbolic regression methods · Ground-truth regression problems, zero target noise | exact solution rate: 53% vs 27% | Helped |
| Use DSR, gplearn, AFP_FE, or AFP instead of AIFeynman at noise levels above 0.01 | Ground-truth equation recovery · Ground-truth regression problems, target noise above 0.01 | exact solution rate | Mixed |
| Use AIFeynman on real-world or noisy data | Symbolic regression · Black-box and noisy regression problems | accuracy | Hurt |
| Use Operon instead of ensemble tree methods | Black-box regression · Black-box regression problems (122 datasets) | model complexity: 1-3 orders of magnitude smaller | Helped |
| Use DSR rather than FFX | Non-GP symbolic regression · Black-box regression problems (122 datasets) | test-set performance | No clear effect |
| Use AIFeynman rather than other methods on Strogatz problems | Ground-truth equation recovery · Strogatz datasets | exact solution rate | No clear effect |
| Add co-evolved fitness estimation to AFP | AFP · Black-box regression problems | method ranking | No clear effect |
| Use epsilon-lexicase selection in FEAT | FEAT · Black-box regression problems | method ranking | No clear effect |
Closest papers by meaning
- Pareto-Optimal Fronts for Benchmarking Symbolic Regression Algorithms (ICML 2025 · similarity 0.91)
- ParFam -- (Neural Guided) Symbolic Regression via Continuous Global Optimization (ICLR 2025 · similarity 0.90)
- SMILE: Bridging Continuous Optimization and Discrete Symbolic Recovery (NeurIPS 2026 · similarity 0.89)
- Ab Initio Nonparametric Variable Selection for Scalable Symbolic Regression with Large $p$ (ICML 2025 · similarity 0.89)
- Symbolic regression via MDLformer-guided search: from minimizing prediction error to minimizing description length (ICLR 2025 · similarity 0.89)
- Improving Monte Carlo Tree Search for Symbolic Regression (NeurIPS 2025 · similarity 0.89)
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.