5 papers
Renet: Principled and Efficient Relaxation for the Elastic Net via Dynamic Objective Selection
Albert Dorador
We introduce Renet, a principled generalization of the Relaxed Lasso to the Elastic Net family of estimators. While, on the one hand, -regularization is a standard tool for…
One Permutation Is All You Need: Fast, Reliable Variable Importance and Model Stress-Testing
Albert Dorador
Reliable estimation of feature contributions in machine learning models is essential for trust, transparency and regulatory compliance, especially when models are proprietary or ot…
Optimizing Feature Ordering in Radar Charts for Multi-Profile Comparison
Albert Dorador
Radar charts are widely used to visualize multivariate data and compare multiple profiles across features. However, the visual clarity of radar charts can be severely compromised w…
TRUST: Transparent, Robust and Ultra-Sparse Trees
Albert Dorador
Piecewise-constant regression trees remain popular for their interpretability, yet often lag behind black-box models like Random Forest in predictive accuracy. In this work, we int…
Theoretical and Empirical Advances in Forest Pruning
Albert Dorador
Regression forests have long delivered state-of-the-art accuracy, often outperforming regression trees and even neural networks, but they suffer from limited interpretability as en…