如何证明概率模型更好
How to show a probabilistic model is better
Washington University in St. Louis（圣路易斯华盛顿大学）
摘要：
我们提出了一个简单的理论框架，和相应的实际程序，在传统的机器学习设置比较真实数据的概率模型。这个框架是基于适当的评分规则的理论，但只需要基本的代数和概率论来理解和验证。所提出的理论概念是经过充分研究的，主要是在统计文献中。本文的目标是倡导他们更广泛地采用经验机器学习的性能评估。
英文摘要：
We present a simple theoretical framework, and corresponding practical procedures, for comparing probabilistic models on real data in a traditional machine learning setting. This framework is based on the theory of proper scoring rules, but requires only basic algebra and probability theory to understand and verify. The theoretical concepts presented are well-studied, primarily in the statistics literature. The goal of this paper is to advocate their wider adoption for performance evaluation in empirical machine learning.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
