Heterogeneity for the Win: One-Shot Federated Clustering

Authors: Don Kurian Dennis, Tian Li, Virginia Smith

ICML 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We motivate our analysis with experiments on common FL benchmarks, and highlight the practical utility of one-shot clustering through usecases in personalized FL and device sampling.
Researcher Affiliation Academia 1Carnegie Mellon University, Pittsburgh, PA, USA. Correspondence to: Don Dennis <dondennis@cmu.edu>.
Pseudocode Yes Algorithm 1 Local kpzq-means (Awasthi & Sheffet, 2012) and Algorithm 2 k-FED are provided with structured steps.
Open Source Code Yes Implementation of k-FED and experimental setup details can be found at: http://github.com/metastable B/kfed/.
Open Datasets Yes We perform this experiment on the FEMNIST and Shakespeare datasets (Caldas et al., 2018) (see Appendix B for details). ... We use the MNIST dataset for this experiment.
Dataset Splits No The paper mentions using datasets like FEMNIST, Shakespeare, and MNIST, but it does not specify explicit train/validation/test split percentages, sample counts, or reference predefined standard splits in detail for reproducibility.
Hardware Specification No The paper describes the context of federated learning involving 'mobile phones or wearables' but does not provide any specific hardware details such as GPU/CPU models, memory, or cloud computing specifications used for running its experiments.
Software Dependencies No The paper does not provide specific software dependency details, such as library or solver names with version numbers, needed to replicate the experiment.
Experiment Setup No The paper states 'Implementation of k-FED and experimental setup details can be found at: http://github.com/metastable B/kfed/' (Section 4), indicating these details are external. The main text itself does not contain specific hyperparameters, optimizer settings, or other detailed system-level training configurations.