Research map / Label-efficient and robust learning / Recognition and long-tail
WeightedSHAP: analyzing and improving Shapley based feature attributions
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 learned weighted marginal contributions instead of uniform Shapley weights | Feature attribution · Synthetic Gaussian (d=100, ρ=0.6) | AUP: 1.65±0.08 -> 0.77±0.03 | Helped |
| Use learned weighted marginal contributions instead of Δd alone | Feature attribution · Synthetic Gaussian (d=100, ρ=0.6) | AUP: 1.49±0.06 -> 0.77±0.03 | Helped |
| Use a weighted mean instead of Δd alone | Feature attribution · Synthetic Gaussian (d=100, ρ=0.6) | estimation error: 9.23±0.02 -> 7.90±0.11 | Helped |
| Use learned weighted marginal contributions instead of uniform Shapley weights | Feature attribution · Synthetic Gaussian (d=100, ρ=0.6) | estimation error: 6.16±0.04 -> 7.90±0.11 | Hurt |
| Rank features using WeightedSHAP instead of Shapley | Gradient-boosted model · airfoil | MSE: 0.53 MSE with 10 features | Helped |
| Rank features using WeightedSHAP instead of Shapley | Gradient-boosted model · fraud | AUC: 0.995 AUC with 14 features | Helped |
| Rank features using WeightedSHAP instead of MCI or Shapley | Gradient-boosted models · Regression datasets | MSE | Helped |
| Rank features using WeightedSHAP instead of MCI or Shapley | Gradient-boosted models · Classification datasets | AUC | Helped |
Closest papers by meaning
- Joint Shapley values: a measure of joint feature importance (ICLR 2022 · similarity 0.91)
- RankSHAP: Shapley Value Based Feature Attributions for Learning to Rank (ICLR 2025 · similarity 0.91)
- shapiq: Shapley Interactions for Machine Learning (NeurIPS 2024 · similarity 0.90)
- A Unified Framework for Provably Efficient Algorithms to Estimate Shapley Values (NeurIPS 2025 · similarity 0.89)
- Provably Accurate Shapley Value Estimation via Leverage Score Sampling (ICLR 2025 · similarity 0.88)
- Unlocking the Game: Estimating Games in Möbius Representation for Explanation and High-Order Interaction Detection (AAAI 2025 · similarity 0.87)
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.