How good Neural Networks interpretation methods really are? A quantitative benchmark

no code implementations • 5 Apr 2023 • Antoine Passemiers, Pietro Folco, Daniele Raimondi, Giovanni Birolo, Yves Moreau, Piero Fariselli

For this purpose, we built synthetic datasets with nonlinearly separable classes and increasing number of decoy (random) features, illustrating the challenge of FS in high-dimensional settings.

feature selection