User-Centric Federated Learning: Trading off Wireless Resources for Personalization
This work addresses the challenge of personalization in Federated Learning for communication-constrained systems, offering an incremental improvement over existing methods.
The paper tackles the problem of statistical heterogeneity in Federated Learning, which increases convergence time and reduces generalization, by designing user-centric aggregation rules based on gradient information to produce personalized models for each client, resulting in improved average accuracy, worst node performance, and reduced training communication overhead compared to baselines.
Statistical heterogeneity across clients in a Federated Learning (FL) system increases the algorithm convergence time and reduces the generalization performance, resulting in a large communication overhead in return for a poor model. To tackle the above problems without violating the privacy constraints that FL imposes, personalized FL methods have to couple statistically similar clients without directly accessing their data in order to guarantee a privacy-preserving transfer. In this work, we design user-centric aggregation rules at the parameter server (PS) that are based on readily available gradient information and are capable of producing personalized models for each FL client. The proposed aggregation rules are inspired by an upper bound of the weighted aggregate empirical risk minimizer. Secondly, we derive a communication-efficient variant based on user clustering which greatly enhances its applicability to communication-constrained systems. Our algorithm outperforms popular personalized FL baselines in terms of average accuracy, worst node performance, and training communication overhead.