Research map / Label-efficient and robust learning / Recognition and long-tail
Decomposing Global Feature Effects Based on Feature Interactions
Julia Herbinger, Marvin N. Wright, Thomas Nagler, Bernd Bischl, Giuseppe Casalicchio
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 |
|---|---|---|---|
| partition the feature space at x3 (GADGET split) before computing regional PD | partial dependence plots · Simulation with feature interaction x1 x x3 | interaction-related heterogeneity: 98% | Improved |
| use ALE instead of PD within GADGET under correlated features | GADGET · Simulation with X1=X3+delta | extrapolation | Improved |
| fit GAMs on Shapley values (SD) instead of PD within GADGET under correlated features | GADGET · Simulation with X1=X3+delta | extrapolation | Mixed |
| increase correlation rho13 between X1 and X3 from 0, 0.4, 0.7 | GADGET · Simulation, 1000 observations | split feature stability: I_3 between 0.75 and 1 | Mixed |
| permute target to build null distribution for PINT | PINT · Type I error theory | type I error | Improved |
Closest papers by meaning
- Regional Explanations: Bridging Local and Global Variable Importance (NeurIPS 2025 · similarity 0.85)
- Sparse and Faithful Local Explanations with Piecewise Linear Surrogates (ICML 2026 · similarity 0.83)
- Exploring the cloud of feature interaction scores in a Rashomon set (ICLR 2024 · similarity 0.83)
- Discovering Features with Synergistic Interactions in Multiple Views (ICML 2024 · similarity 0.82)
- Trimming the Long-Tail of Visual World Modeling Evaluation (NeurIPS 2026 · similarity 0.82)
- Fast Estimation of Partial Dependence Functions using Trees (ICML 2025 · similarity 0.81)
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.
