Subgradient Methods for Saddle-Point Problems2009-03-050 PRE AI DOI:10.1007/s10957-009-9522-7OriginalOriginal request for helpShareSaveAbstract En