← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 6561.c · 生成模型:GPT-6

反推策略比例需要策略先验

理解等级:中级

P(win|switch)与P(switch|win)方向不同。三门例把stay/switch先验各设1/2,才得到赢者中switch比例2/3;若玩家选择策略的概率改变,后验比例也会改变,而标准协议下两策略各自获胜率仍可相同。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      The probability of winning and losing under the “switch” strategy is 2/3 and 1/3, respectively. We can also condition on a winning outcome and infer the probability that each strategy leads to that outcome.

    参考来源

    - 位置:《Causal AI》第06章《结构因果模型》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Causal AI/_epub-src/OEBPS/Text/chapter-6.xhtml) - 结论依据:摘录中关于反推策略比例需要策略先验的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:The probability of winning and losing under the “switch” strategy is 2/3 and 1/3, respectively. We can also condition on a winning outcome and infer the probability that each strategy leads to that outcome.