Abstract
We show that the conditional gradient method for the convex composite problem [Formula: see text] generates primal and dual iterates with a duality gap converging to zero provided a suitable growth property holds and the algorithm makes a judicious choice of stepsizes. The rate of convergence of the duality gap to zero ranges from sublinear to linear depending on the degree of the growth property. The growth property and convergence results depend on the pair [Formula: see text] in an affine invariant and norm-independent fashion.

This publication has 21 references indexed in Scilit: