Unmanned aerial vehicle (UAV)-assisted networks, where small base stations are mounted to one or more UAVs, are envisioned as a viable method to provide seamless and ubiquitous wireless connectivity. The UAV-assisted wireless networks have a wide variety of applications, including public safety, disaster management, and many others. Due to the ‘fluid’ nature of the coverage area in UAV-assisted networks, the deployment of multiple UAVs can cause significant interference to users at the cell edge. To ensure adequate throughput for these cell-edge users, network resources such as bandwidth and transmit power must be intelligently allocated among the various UAVs. In this paper, we propose a Super Learning-based resource allocation scheme for UAV-assisted wireless networks. To this end, we introduce a super learner model that allocates resource plans of soft frequency reuse (SFR) to various UAVs. The proposed scheme significantly improves the accuracy of allocating the resource plans among various UAVs. Without loss of generality, we consider the deployment of a total of eight UAVs. The proposed super learner model combines 15 classifiers. Simulation results show that the proposed technique provides 75.85% accuracy with an improvement of 3.8% by using 80% fewer data compared to the existing work.