**5. Conclusion**

Artificial intelligence is becoming increasingly important for enhancing people's quality of life and boosting productivity. Artificial intelligence is becoming increasingly important for enhancing people's quality of life and boosting productivity. The integration of edge computing with Federated Learning (FL) can help to tackle the data privacy issue. However, federated learning involves a significant amount of training overhead, which can be a challenge for resource-limited end devices. We propose a solution to reduce the system overhead of FL and make it more affordable to edge computing by automatically adjusting FL hyper-parameters. Our preliminary work has demonstrated promising results, with up to 26% reduction in system overhead. This suggests that FL hyper-parameter tuning is an effective approach for edge computing. However, further research is needed to fully support FL in edge computing, and more applications are required to drive the growth of the edge computing ecosystem.
