Angel REYERO LOBO — accepted AI-conference papers

2026

Aggregate Models, Not Explanations: Improving Feature Importance Estimation

Joseph Paillard, Angel REYERO LOBO, Denis-Alexander Engemann, Thirion Bertrand

ICML 2026poster

Feature-importance methods show promise for transforming machine learning (ML) models from predictive engines into tools for scientific discovery. However, expressive models can be unstable due to data sampling and algorithmic stochasticity, leading to inaccurate variable importance estimates, under…

Cited by 1 Source ↗Scholar ↗

2026

Semi-knockoffs: a model-agnostic conditional independence testing method with finite-sample guarantees

Angel REYERO LOBO, Thirion Bertrand, Pierre Neuvial

ICML 2026poster

Conditional independence testing (CIT) is essential for reliable scientific discovery. It prevents spurious findings and enables controlled feature selection. Recent CIT methods have used machine learning (ML) models as surrogates of the underlying distribution. However, model-agnostic approaches re…

Cited by 0 Source ↗Scholar ↗