Research map / Learning theory and algorithms / Prediction and decision losses
Model Class Reliance for Random Forests
Gavin Smith, Roberto Mansilla, James Goulding
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 |
|---|---|---|---|
| add prediction-equivalent tree substitution on top of surrogate splits (ablation) | RF-MCR+ · COMPAS / Cancer | MCR+ increase: surrogates account for 1.11 −15.38% of increase; rest from tree equivalence | Improved |
| add prediction-equivalent tree substitution on top of surrogate splits (ablation) | RF-MCR- · COMPAS / Cancer | MCR- decrease: surrogates account for 2.06−100% of decrease; rest from tree equivalence | Improved |
| use RF-MCR instead of INFL for upper bound | MCR+ estimation · Synthetic XOR dataset | MCR+ recovery | Improved |
| use RF-MCR instead of approximate RF-cPI for lower bound | MCR- estimation · Synthetic XOR dataset | MCR- recovery | Improved |
| shrink Rashomon set epsilon from 3.17 to 0 (ablation) | RF-MCR · COMPAS (age) | MCR range | Hurt |
Closest papers by meaning
- All Models are Wrong, but Many are Useful: Learning a Variable's Importance by Studying an Entire Class of Prediction Models Simultaneously (JMLR 2019 · similarity 0.90)
- From global to local MDI variable importances for random forests and when they are Shapley values (NeurIPS 2021 · similarity 0.85)
- The Rashomon Importance Distribution: Getting RID of Unstable, Single Model-based Variable Importance (NeurIPS 2023 · similarity 0.84)
- Identifying a Minimal Class of Models for High--dimensional Data (JMLR 2017 · similarity 0.84)
- A Unified Framework for Provably Efficient Algorithms to Estimate Shapley Values (NeurIPS 2025 · similarity 0.82)
- Random features models: a way to study the success of naive imputation (ICML 2024 · similarity 0.82)
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.
