Robust optimization equilibria for bimatrix game

In this paper, we investigate robust optimization equilibria with two players, in which each player can not evaluate his own cost matrix accurately while may estimate a bounded set of the cost matrix. We obtain a result that solving this equilibria can be formulated as solving a second-order cone complementarity problem under l(2)-norm or a mixed complementarity problem under l(1)boolean AND l(infinity)-norm. We present some numerical results to illustrate the relation between robustness and optimality.