1: \begin{abstract}\vspace{-2mm}

2: %

3: We study the linear convergence of variants of the Frank-Wolfe algorithms for some classes of strongly convex problems, using only affine-invariant quantities. 

4: As in~\cite{Guelat:1986fq}, we show the linear convergence of the standard Frank-Wolfe algorithm when the solution is in the interior of the domain, but with affine invariant constants. We also show the linear convergence of the away-steps variant of the Frank-Wolfe algorithm, but with constants which only depend on the geometry of the domain, and not any property of the location of the optimal solution.

5: Running these algorithms does not require knowing any problem specific parameters.

6: %

7: %

8: \end{abstract}

9: