BioMutFed+: Reproducibility Artifacts (TMC-2025-09-2616)

Le résumé fourni par la source

Reproducibility artifacts for "BioMutFed+: Mutation-Driven Federated Learning for IIoT," IEEE Transactions on Mobile Computing (TMC-2025-09-2616). Two Jupyter notebooks reproduce the ten-seed Wisconsin Breast Cancer evaluation (Tables VI and VII) and the lambda sensitivity sweep within the full BioMutFed+ pipeline (Supplementary Section SIII-F, Table S2). All experiments use Dirichlet alpha=0.1 partitioning across 20 clients with 20% gradient-ascent adversarial clients. Each notebook is self-contained and produces the exact numerical values reported in the paper. See README.md for environment details and runtime.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.