Pre-registration of a replication experiment for the study 'Grouping or Ordering? Coverage-Aware Sequential Federated Learning under Label Skew'. The plan (REPLICATION_PLAN.md), the analysis script and the job grids were fixed at commit eab3876c1239249223e9864f35d4744294aa5b97 of the study repository before any replication run; this record timestamps them. Primary hypothesis: FedDAC-Order exceeds random grouping with random chain order in final test accuracy (exact two-sided Wilcoxon, Holm over three settings). Settings: CIFAR-100 with ResNet-18 (N = 50), FEMNIST with its natural partition by writer (N = 50 and N = 100). Code and results of the study: Zenodo DOI 10.5281/zenodo.22965988.
No takes yet. Share an insight, caveat, or question.
Salah et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: