arXiv · Statistics Machine Learning· JM Gorriz·· 1 天前AI 评分27
准确率何时才是证据？泛化、验证与信息融合的统一理论
When Is Accuracy Evidence? A Unified Theory of Generalisation, Validation, and Information Fusion
AI 导读
论文提出 gamma-CUBV 框架，用指数族矩生成函数与累积量包络 gamma(lambda) 统一控制泛化间隙，给出覆盖 Hoeffding、Bernstein、依赖感知、PAC-Bayesian 与异质源融合的风险上界。
来源：arXiv · Statistics Machine Learning · arxiv.org