Skip to content

Commit

Permalink
update README
Browse files Browse the repository at this point in the history
  • Loading branch information
halfrost committed Mar 5, 2018
1 parent 837a901 commit 38fa047
Showing 1 changed file with 3 additions and 3 deletions.
6 changes: 3 additions & 3 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -381,7 +381,7 @@ E [L(T)] = 20 * 0.25)+(30 * 0.5)+(60 * 0.25)= 35

得到这 16 种情况以后,接着继续往下递归。递归公式如下:

![](http://latex.codecogs.com/gif.latex?\score(G)=\\sum%20E(G)=%20\\sum%20P(G',G)%20*%20MAX%20(score(G%20\\rightarrow%20D)))
![](http://latex.codecogs.com/gif.latex?score(G)%20=%20\\sum%20E(G)=%20\\sum%20P(G',G)%20*%20MAX%20(score(G%20\\rightarrow%20D)))

上面公式就是不断进行期望值的计算。

Expand All @@ -390,7 +390,7 @@ E [L(T)] = 20 * 0.25)+(30 * 0.5)+(60 * 0.25)= 35

递归收敛以后,就开始计算本次的期望,这个期望值是由权重矩阵和棋盘矩阵相乘得到的值。权重矩阵里面的值也是需要自己调整的,调整的不好会导致递归层次很多,影响效率;递归层次太少,又会影响期望结果计算的准确性。这个权重矩阵的“调教”也许可以交给机器学习的无监督学习去做。

![](http://latex.codecogs.com/gif.latex?\score(G)%20=%20\\sum%20E(G)%20=%20\\sum%20P(G',G)%20*%20MAX%20(%20\\prod\\left%20\\|%20W*G'%20\\right%20\\|%20))
![](http://latex.codecogs.com/gif.latex?score(G)%20=%20\\sum%20E(G)%20=%20\\sum%20P(G',G)%20*%20MAX%20(%20\\prod\\left%20\\|%20W*G'%20\\right%20\\|%20))

上述公式就是递归收敛条件下的期望值计算公式。

Expand All @@ -399,7 +399,7 @@ E [L(T)] = 20 * 0.25)+(30 * 0.5)+(60 * 0.25)= 35
和上面举的去机场的例子一样,计算好各条路线的期望值。这里计算好最大期望值以后,再求一个均值就好了,值最大的就是下一步需要移动的方向。


![](http://latex.codecogs.com/gif.latex?\Dir(G)%20=%20AVG%20\\cdot%20MAX%20(score(G%20\\rightarrow%20D)))
![](http://latex.codecogs.com/gif.latex?Dir(G)%20=%20AVG%20\\cdot%20MAX%20(score(G%20\\rightarrow%20D)))


不过在实际递归过程是会出现下面这种情况:
Expand Down

0 comments on commit 38fa047

Please sign in to comment.