The Exact Feasibility of Randomized Solutions of Uncertain Convex Programs

摘要: Many optimization problems are naturally delivered in an uncertain framework, and one would like to exercise prudence against the uncertainty elements present problem. In previous contributions, it has been shown that solutions convex programs bear a high probability satisfy constraints can be obtained at low computational cost through constraint randomization. this paper, we establish new feasibility results for randomized algorithms. Specifically, exact class of so-called fully-supported is obtained. It turns out all share same properties, revealing deep kinship among class. further proven other bounded based on prototype problems. The result paper outperforms bounds not improvable because