Orthogonal variance decomposition based feature selection
This addresses feature selection for machine learning practitioners, but appears incremental as it builds on existing variance decomposition approaches.
The paper tackled the problem of feature selection methods failing to account for feature interactions by using orthogonal variance decomposition to evaluate features, resulting in an efficient algorithm that improved model accuracy in numerical experiments.
Existing feature selection methods fail to properly account for interactions between features when evaluating feature subsets. In this paper, we attempt to remedy this issue by using orthogonal variance decomposition to evaluate features. The orthogonality of the decomposition allows us to directly calculate the total contribution of a feature to the output variance. Thus we obtain an efficient algorithm for feature evaluation which takes into account interactions among features. Numerical experiments demonstrate that our method accurately identifies relevant features and improves the accuracy of numerical models.