Optimization & Bounded QP (Extension)
MED maximizes concave objective with convex constraints.
Axis-parallel, Newton, gradient descent will converge.
Lower Bound the concave objective with quadratic
Can then use SMO, QP, and other SVM optimizers.
Example: SVM Feature Selection
Iterate bound (contact at ) and QP
Each QP is seeded at previous solín
Converges in about 10 fast iterations