知识卡片
奥卡姆偏好不证明简单即真
多个规则同样解释有限样本时,简单性可以作为额外选择偏好,但简单的定义依赖表示与先验。更大空间分散先验质量的直觉不自动证明任意简单假设更可能为真,尤其连续参数中需区分密度与点概率;偏好还应接受新证据检验。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 泛化问题
原书出处 · 1 处
- · 查看原书依据
We suspect your reason is again Occam's razor, which states that when multiple hypotheses equally well fit the data, you should prefer the simplest. To a human, it may be that x ⋆ y = (xy)2 is the simplest. To a computer, defining a proper notion of simplicity is a hard problem, but can be made precise.
参考来源
- 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第11章《泛化问题》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)/_epub-src/OPS/xhtml/c011.xhtml)
- 结论依据:摘录中关于奥卡姆偏好不证明简单即真的定义、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:We suspect your reason is again Occam's razor, which states that when multiple hypotheses equally well fit the data, you should prefer the simplest. To a human, it may be that x ⋆ y = (xy)2 is the simplest. To a computer, defining a proper notion of simplicity is a hard problem, but can be made precise.