摘要 : A p-norm surrogate constraint method is proposed for integer programming. A single surrogate constraint can be always constructed using a p-norm such that the feasible sets in a surrogate relaxation and the primal problem match ex... 展开 A p-norm surrogate constraint method is proposed for integer programming. A single surrogate constraint can be always constructed using a p-norm such that the feasible sets in a surrogate relaxation and the primal problem match exactly. The p-norm surrogate constraint method is thus guaranteed to succeed in identifying an optimal solution of the primal problem with zero duality gap. The existence of a saddle point is proven for integer programming problems. 收起