决策充分后验近似论文发布

1. 12:00
   arXiv · Statistics Machine Learning

   决策充分的后验近似：理论框架与有效维度

   论文提出决策充分的后验近似框架，要求后验近似保留特定下游决策问题的结构。目标后验与损失函数定义动作上的遗憾几何，基线近似决定诱导动作变化所需的 forward-Kullback-Leibler 信息，受限近似族决定可用的变化；KL 散度在 Bayes-action fibers 上的收缩给出到决策充分与决策失败的精确距离。

Current heat 9·Comparable peak 10（Oct 9）·Comparable change over 24 hours –

The trend compares only the same participants observed continuously; its range may be smaller than the current heat count. Move or click on the chart to inspect hourly heat; use the left and right arrow keys to switch.