Search papers, labs, and topics across Lattice.
This paper introduces a pre-deployment framework for estimating the complexity of federated learning tasks in edge perception systems. The framework combines intrinsic data properties (dimensionality, sparsity, heterogeneity) with client composition characteristics to produce a complexity metric. Experiments on MNIST and CIFAR variants demonstrate a strong correlation between the proposed metric and both federated learning performance and communication costs.
Save time and resources: predict federated learning performance *before* deployment by quantifying dataset and client complexity.
Edge AI systems increasingly rely on federated learning to train perception models in distributed, privacy-preserving, and resource-constrained environments. Yet, before training begins, practitioners often lack practical tools to estimate how difficult a federated learning task will be in terms of achievable accuracy and communication cost. This paper presents a classifier-agnostic, pre-deployment framework for estimating learning complexity in federated perception systems by jointly modeling intrinsic properties of the data and characteristics of the distributed environment. The proposed complexity metric integrates dataset attributes such as dimensionality, sparsity, and heterogeneity with factors related to the composition of participating clients. Using federated learning as a representative distributed training setting, we examine how learning difficulty varies across different federated configurations. Experiments on multiple variants of the MNIST dataset and CIFAR dataset show that the proposed metric strongly correlates with federated learning performance and the communication effort required to reach fixed accuracy targets. These findings suggest that complexity estimation can serve as a practical diagnostic tool for resource planning, dataset assessment, and feasibility evaluation in edge-deployed perception systems.
