AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Statistically Valid Variable Importance Assessment through Conditional Permutations

Ahmad Chamma, Denis Engemann, Bertrand Thirion

NeurIPS 2023 · Probabilistic and scientific modeling › Molecules and proteins · approach: Statistical analysis of ML practice · 6 citations (OpenAlex)

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 changedWhereMeasuredEffect
Use conditional permutation instead of standard permutationDNN variable importance · Simulations with increasing variable correlationtype-I error controlImproved
Use conditional permutation instead of standard permutationDNN variable importance · Simulations with increasing variable correlationAUCNo clear effect
Use conditional permutation instead of standard permutationDNN variable importance · Five simulated data-generating scenariostype-I errorImproved
Use conditional permutation instead of standard permutationDNN variable importance · Five simulated data-generating scenariosAUCImproved
Use conditional permutation instead of standard permutationDNN variable importance · UK Biobank-based simulationAUCImproved
Use conditional permutation instead of standard permutationDNN variable importance · UK Biobank-based simulationpowerImproved
Use conditional permutation instead of standard permutationDNN variable importance · UK Biobank-based simulationtype-I errorNo clear effect
Use conditional permutation instead of standard permutationDNN variable importance · Large-scale simulation, correlation coefficient 0.8type-I errorImproved
Use conditional permutation instead of standard permutationDNN variable importance · UK Biobank age predictionselected brain inputs: 16 out of 66Mixed
Use conditional permutation instead of standard permutationDNN variable importance · Simulations across data-generating scenariospowerNo clear effect
Use conditional permutation instead of standard permutationDNN variable importance · High-correlation simulationscomputation timeHurt
Use conditional permutation with an RF conditional model rather than DNN as the importance learnerCPI variable importance · Simulated scenariosAUCHurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy