[PDF] Efficient Data Distribution Estimation for Accelerated Federated Learning | Semantic Scholar (2024)

[PDF] Semantic Reader

HACCS is a Heterogeneity-Aware Clustered Client Selection system that identifies and exploits the statistical heterogeneity by representing all distinguishable data distributions instead of individual devices in the training process, and can provide 18% −38% reduction in time to convergence compared to the state of the art.

This article introduces a novel client selection framework that judiciously leverages correlations across local datasets to accelerate training and designs a novel Neural Contextual Combinatorial Bandit algorithm to establish relationships between client features and rewards, enabling intelligent selection of client combinations.

An adaptive FMTL framework is developed, which can accelerate the model training convergence and reduce the computation complexity while ensuring model accuracy, and is validated in the edge learning model.

This paper critically reviews recent CS methods for FL and analyses the CS methods, their functionality, and their limitations, and provides a comparison of the used approaches in terms of how they are evaluated.

This paper introduces a multiscale algorithmic framework which combines theoretical guarantees of FedAvg and FedOMD algorithms in near stationary settings with a non-stationary detection and adaptation technique to ameliorate FL generalization performance in the presence of concept drifts.

This article systematically presents recent advances in the emerging field of FL client selection and its challenges and research opportunities to facilitate practitioners in choosing the most suitable client selection mechanisms for their applications, as well as inspire researchers and newcomers to better understand this exciting research topic.

A novel sampling method named the irrelevance sampling technique that selects a subset of clients based on quality and quantity of data on edge devices that achieves 50–80% faster convergence even in highly skewed data distribution in the presence of free riders.

FedScale is presented, a federated learning benchmarking suite with realistic datasets and a scalable runtime to enable reproducible FL research and highlight potential opportunities for heterogeneity-aware co-optimizations in FL.

It is proved that clustered sampling leads to better clients representatitivity and to reduced variance of the clients stochastic aggregation weights in FL and is compatible with existing methods and technologies for privacy enhancement, and for communication reduction through model compression.

This work analyses the impact of data heterogeneity on device selection, model convergence, model accuracy, and fault tolerance in a federated learning setting and proposes three methods for identifying groups of devices with similar data distributions that can significantly improve the model convergence without compromising model accuracy.

...

...