TabNet
A deep learning model for tabular data introduced by Google in 2019. It employs an attention mechanism for sequential feature selection, aiming to replace gradient boosting trees with neural networks for structured data processing.
Timeline (last 90 days)
Google introduced TabNet in 2019.
Frank Hutter noted that TabNet was once heavily hyped and received thousands of citations, but failed to generalize to new datasets.
A large-scale benchmark study published at NeurIPS in 2022 showed that on small- to medium-scale tabular data tasks, gradient boosting trees outperformed deep learning models (including TabNet and SAINT) in the majority of scenarios across 45 datasets.
All Facts (3)
Frank Hutter noted that TabNet was once heavily hyped and received thousands of citations, but failed to generalize to new datasets.
50%UnverifiedA large-scale benchmark study published at NeurIPS in 2022 showed that on small- to medium-scale tabular data tasks, gradient boosting trees outperformed deep learning models (including TabNet and SAINT) in the majority of scenarios across 45 datasets.
50%UnverifiedGoogle introduced TabNet in 2019.
50%