The generic splicing technique certifiably guarantees the best subset can be selected in a polynomial time. In practice, the computational efficiency can be improved to handle large scale datasets. The tips for computational improvement are applicable for:
ultra-high dimensional data via
focus on important variables;
large-sample data via
sparse inputs via
sparse matrix computation;
specific models via
covariance update for
quasi Newton iteration for
More importantly, the technique in these tips can be use simultaneously.
abess allow algorithms to use both feature screening and golden-section searching such that
algorithms can handle datasets with large-sample and ultra-high dimension.
The following contents illustrate the above tips.
abess efficiently implements warm-start initialization and parallel computing,
which are very useful for fast computing.
To help use leverage them, we will also describe their implementation details in the following.