d*****n
发帖数: 39 | 1 题目感觉不难，但是突然被问到，结果从variance和covariance上绕了半天也没搞好，
感觉面试官对最后的答案也不满意。
题目： Drawing a pair of (x, y) from a joint Gaussian distribution with 0
covariance. Knowing the standard deviations of x and y and knowing z = x + y
, what is your best guess for x?
求牛人给个解释 |
E**********e
发帖数: 1736 | 2 is it typo, you need guess z or x? |
d***y
发帖数: 1 | 3 Check linear minimum mean square error estimator (MMSE for Gaussian).
Basically you need to estimate E[x|z], it has a closed form solution when
everything is Gaussian. |
d*****n
发帖数: 39 | 4 Guess X
【在 E**********e 的大作中提到】
: is it typo, you need guess z or x?
|
d*****n
发帖数: 39 | 5 I think it is in the form E[x|z] as well.
【在 d***y 的大作中提到】
: Check linear minimum mean square error estimator (MMSE for Gaussian).
: Basically you need to estimate E[x|z], it has a closed form solution when
: everything is Gaussian.
|
