知识卡片
残差代价与对数似然符号相反
正的eᵀΩe对应负对数似然的一部分,最小化它才与最大化高斯似然一致。原式17.13把正二次项称为对数似然漏负号,17.14的最小化方向则符合加权最小二乘。
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
- · 同时定位与建图
原书出处 · 1 处
- · 查看原书依据
Again, the log-likelihood for observation zij is directly derived from the definition of the normal distribution, but using the information matrix instead of the covariance matrix, it is ridden of the exponential function by taking the logarithm on both sides.
参考来源
- 位置:《Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)》第17章《同时定位与建图》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)/_epub-src/OEBPS/xhtml/chapter_17.xhtml)
- 结论依据:摘录中关于残差代价与对数似然符号相反的定义、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:Again, the log-likelihood for observation zij is directly derived from the definition of the normal distribution, but using the information matrix instead of the covariance matrix, it is ridden of the exponential function by taking the logarithm on both sides.