EPPFL

This dataset is associated with the manuscript titled \u201cA Robust and Efficient Privacy-Preserving Federated Learning Framework Based on Functional Encryption.\u201d It is used to evaluate the performance of the proposed RobEPPFL framework in terms of model accuracy, runtime, and communication overhead. Specifically, two common image classification datasets\u2014MNIST and CIFAR10\u2014were employed. These datasets encompass  different levels of task complexity, ranging from simple grayscale image recognition to color image classification, to test the performance of the proposed framework across a variety of difficulty levels. Using these datasets, experiments validated the superiority of the proposed framework over existing methods in terms of model accuracy, computational efficiency, and communication efficiency.