A key problem of component-based grid application configuration is to map services onto the execution nodes of the grid environment such that all services of the application satisfy some minimum quality requirements. This problem is known to be NP-hard. Based on the previous ant-based methods to solve the problem, this paper proposes learning gatekeepers to guide the movement of unintelligent foraging ants to enhance the efficiency of the grid node graph exploration. In this respect, three variants of learning schemes have been used to reason how the gatekeepers could learn better recommendations to guide the ants. The simulation evaluation of the proposed methods shows that inclusion of gatekeepers significantly improves the convergence performance over the simple unintelligent random ants.