Credit-to-Ranking Alignment for Ordered Task Allocation in Hierarchical Multi-Agent Reinforcement Learning for Vehicular Edge Computing: Reproducibility Package

Résumé fourni par la source

This repository provides the reproducibility package for the study “Credit-to-Ranking Alignment for Ordered Task Allocation in Hierarchical Multi-Agent Reinforcement Learning for Vehicular Edge Computing.” It contains the frozen confirmatory state registry, factorial and supervised evaluation records, state-level metrics, statistical analysis plan, derived tables, crossed panel–seed–state bootstrap outputs, and executable analysis scripts used to reproduce the confirmatory statistical results reported in the manuscript. The package reproduces the principal factorial contrasts, ranking analyses, state-registry separation checks, and selected figures from the frozen evaluation data. It is intended to reproduce the confirmatory evaluation and statistical analyses; it does not constitute a complete end-to-end training package for retraining all planner variants from scratch.

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