SparsePCA plus XGBoost can generate synthetic tabular data and latent-space perturbations that serve as an interpretable alternative to raw and quantile noise for robustness testing.
As expected, the 3D point clouds are linearly projected on a plane 7 FIG
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Towards a framework on tabular synthetic data generation: a minimalist approach: theory, use cases, and limitations
SparsePCA plus XGBoost can generate synthetic tabular data and latent-space perturbations that serve as an interpretable alternative to raw and quantile noise for robustness testing.