SLEP: A Sparse Learning Package

[Home] [Overview] [Feature] [Download] [Manual] [Citation]

The SLEP package has the following main features:

First-Order Method

At each iteration, we only need to evaluate the function value and the gradient; and thus the algorithms can handle large-scale sparse data.

Optimal Convergence Rate

The convergence rate O(1/k²) is optimal for smooth convex optimization via the first-order black-box methods.

Efficient Projection

The projection problem can be solved efficiently.

Pathwise Solutions

The SLEP package provides functions that can efficiently compute the pathwise solutions corresponding to a series of regularization parameters using the “warm-start” technique.