arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.01573stat.MEstat.ML
收藏

何时才是广义贝叶斯的贝叶斯?一种基于损失函数的决策理论对损失更新的特征化

When Is Generalized Bayes Bayesian? A Decision-Theoretic Characterization of Loss-Based Updating

Temple University(天普大学);RIKEN Center for Advanced Intelligence Project(理化学研究所先进智能项目中心)

Kenichiro McAlinn, Kōsaku Takanashi

AI总结:

本文通过决策理论特征化,区分了信念后验与决策后验,并证明基于损失的后验在特定条件下等同于普通贝叶斯,同时揭示了决策后验需要非线性偏好及熵惩罚的变分表示。

摘要:

基于损失函数的更新,包括广义贝叶斯、吉布斯和准后验,通过用户选择的损失函数替代似然函数,并通过指数倾斜产生后验分布。我们提供了一种决策理论的特征化,将“信念后验”——由Savage和Anscombe-Aumann的联合概率模型基础所支持的条件信念,与“决策后验”——由决策规则偏好所支持的随机决策规则区分开来。我们明确指出,基于损失的后验仅在损失是(除尺度和数据相关项外)负对数似然时才与普通贝叶斯相同。然后我们展示广义边缘似然不是决策后验的证据,并且在没有额外结构的情况下,贝叶斯因子并不明确。在决策后验领域,非退化的后验需要非线性偏好。在顺序一致性与分离性下,这些偏好导致熵惩罚的变分表示,从而产生广义贝叶斯作为最优规则。

英文摘要:

Loss-based updating, including generalized Bayes, Gibbs, and quasi-posteriors, replaces likelihoods by a user-chosen loss and produces a posterior-like distribution via exponential tilt. We give a decision-theoretic characterization that separates \emph{belief posteriors} -- conditional beliefs justified by the foundations of Savage and Anscombe-Aumann under a joint probability mode l-- from \emph{decision posteriors} -- randomized decision rules justified by preferences over decision rules. We make explicit that a loss-based posterior coincides with ordinary Bayes if and only if the loss is, up to scale and a data-only term, negative log-likelihood. We then show that generalized marginal likelihood is not evidence for decision posteriors, and Bayes factors are not well-defined without additional structure. In the decision posterior regime, non-degenerate posteriors require nonlinear preferences over decision rules. Under sequential coherence and separability, these lead to an entropy-penalized variational representation yielding generalized Bayes as the optimal rule.

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。

↑