Shakespeare

The 'Shakespeare' dataset is a benchmarking dataset used to evaluate the performance of Federated Learning algorithms in terms of accuracy, convergence time, communication overhead, energy consumption, and robustness to non-IID data.

Papers using Shakespeare (15)

Benchmarking Federated Learning in Edge Computing Environments: A Systematic Review and Performance Evaluation2026FedZMG: Efficient Client-Side Optimization in Federated Learning2026CLIP: Client-Side Invariant Pruning for Mitigating Stragglers in Secure Federated Learning2025TurboSVM-FL: Boosting Federated Learning through SVM Aggregation for Lazy Clients2024 · 2 citesFederated Hyperparameter Tuning: Challenges, Baselines, and Connections to Weight-Sharing2021 · 12 citesDifferentially Private Federated Learning with Laplacian Smoothing2020 · 9 citesAggregate or Not? Exploring Where to Privatize in DNN Based Federated Learning Under Different Non-IID Scenes2021 · 8 citesSubject Granular Differential Privacy in Federated Learning2022 · 6 citesFederated Learning Under Intermittent Client Availability and Time-Varying Communication Constraints2022 · 3 citesFedBC: Calibrating Global and Local Models via Federated Learning Beyond Consensus2022 · 3 citesCANIFE: Crafting Canaries for Empirical Privacy Measurement in Federated Learning2022 · 3 citesFederated Asymptotics: a model to compare federated learning algorithms2021 · 2 citesCommunication-Efficient Federated Learning via Optimal Client Sampling2020 · 1 citesAccelerating Federated Learning with a Global Biased Optimiser2021 · 1 citesFaster Federated Learning with Decaying Number of Local SGD Steps2023