mlsanityData to 7 Oct 2026

Decorrelated Variable Importance

arXiv 2111.10853 ↗First posted 21 Nov 2021stat.ME stat.MLcited by 12

Not on the map. Why?

Abstract (arXiv)

Because of the widespread use of black box prediction methods such as random forests and neural nets, there is renewed interest in developing methods for quantifying variable importance as part of the broader goal of interpretable prediction. A popular approach is to define a variable importance parameter - known as LOCO (Leave Out COvariates) - based on dropping covariates from a regression model. This is essentially a nonparametric version of R-squared. This parameter is very general and can be estimated nonparametrically, but it can be hard to interpret because it is affected by correlation between covariates. We propose a method for mitigating the effect of correlation by defining a modified version of LOCO. This new parameter is difficult to estimate nonparametrically, but we show how to estimate it using semiparametric models.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Builds on 14 of its 33 references that are in the map. First posted on arXiv 21 Nov 2021.

Builds on

14 of 33 references.

References · 14

  1. 2016Distribution-Free Predictive Inference For Regressioncited by 657
  2. 2016Model-Agnostic Interpretability of Machine Learningcited by 219
  3. 2019Explaining individual predictions when features are dependent: More accurate approximations to Shapley valuescited by 205
  4. 2020Improving KernelSHAP: Practical Shapley Value Estimation via Linear Regressioncited by 153
  5. 2020Efficient nonparametric statistical inference on population feature importance using Shapley valuescited by 61
  6. 2019Explaining black box decisions by Shapley cohort refinementcited by 26
  7. 2021SHAFF: Fast and consistent SHApley eFfect estimates via random Forestscited by 22
  8. 2016An unexpected unity among methods for interpreting model predictionscited by 16
  9. 2020Floodgate: inference for model-free variable importancecited by 15
  10. 2021Significance tests of feature relevance for a black-box learnercited by 14
  11. 2018Robust Estimation of Causal Effects via High-Dimensional Covariate Balancing Propensity Scorecited by 12
  12. 2020A Semiparametric Approach to Interpretable Machine Learningcited by 3
2 more
  1. 2021Variable importance scorescited by 2
  2. 2020Shapley value confidence intervals for attributing variance explainedcited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 3; 2023: 1; 2024: 4; 2025: 3; 2026: 1.

12 papers.

  1. 2023Feature Importance: A Closer Look at Shapley Values and LOCOcited by 23
  2. 2022Scientific Inference With Interpretable Machine Learning: Analyzing Models to Learn About Real-World Phenomenacited by 21
  3. 2024Rethinking Explainable Machine Learning as Applied Statisticscited by 8
  4. 2022SHAP-XRT: The Shapley Value Meets Conditional Independence Testingcited by 6
  5. 2024I Bet You Did Not Mean That: Testing Semantic Importance via Bettingcited by 3
  6. 2022Shapley Curves: A Smoothing Perspectivecited by 2
  7. 2024Conditional Mean and Variance Estimation via textitk-NN Algorithm with Automated Variance Selectioncited by 2
  8. 2025iLOCO: Distribution-Free Inference for Feature Interactionscited by 2
  9. 2025Decorrelated feature importance from local sample weightingcited by 1
  10. 2026MinShap: A Shapley-Based Framework for Feature Redundancycited by 1
  11. 2024Targeted Learning for Variable Importancecited by 0
  12. 2025AICO: Feature Significance Tests for Supervised Learningcited by 0