要从联合概率中得到条件概率,我们可以使用条件概率的定义。具体来说,给定两个随机变量 x x x 和 y y y,其联合概率 P ( x , y ) P(x, y) P(x,y),条件概率 P ( x ∣ y ) P(x|y) P(x∣y) 和 P ( y ∣ x ) P(y|x) P(y∣x) 可以通过以下方式得到:
条件概率的定义
-
条件概率 p ( y ∣ x ) p(y|x) p(y∣x):
条件概率 p ( y ∣ x ) p(y|x) p(y∣x) 表示在给定 x x x 的条件下, y y y 的概率。它可以通过联合概率和边缘概率计算得到:
p ( y ∣ x ) = p ( x , y ) p ( x ) p(y|x) = \frac{p(x, y)}{p(x)} p(y∣x)=p(x)p(x,y)
其中 p ( x ) p(x) p(x) 是 x x x 的边缘概率,计算方式如下:
p ( x ) = ∑ y p ( x , y ) (对于离散情况) p ( x ) = ∫ p ( x , y ) d y (对于连续情况) p(x) = \sum_y p(x, y) \quad \text{(对于离散情况)} \\ p(x) = \int p(x, y) \, dy \quad \text{(对于连续情况)} p(x)=y∑p(x,y)(对于离散情况)p(x)=∫p(x,y)dy(对于连续情况) -
条件概率 p ( x ∣ y ) p(x|y) p(x∣y):
条件概率 p ( x ∣ y ) p(x|y) p(x∣y) 表示在给定 y y y 的条件下, x x x 的概率。它可以通过联合概率和边缘概率计算得到:
p ( x ∣ y ) = p ( x , y ) p ( y ) p(x|y) = \frac{p(x, y)}{p(y)} p(x∣y)=p(y)p(x,y)
其中 p ( y ) p(y) p(y) 是 y y y 的边缘概率,计算方式如下:
p ( y ) = ∑ x p ( x , y ) (对于离散情况) p ( y ) = ∫ p ( x , y ) d x (对于连续情况) p(y) = \sum_x p(x, y) \quad \text{(对于离散情况)} \\ p(y) = \int p(x, y) \, dx \quad \text{(对于连续情况)} p(y)=x∑p(x,y)(对于离散情况)p(y)=∫p(x,y)dx(对于连续情况)
示例
假设我们有以下联合概率:
p
(
x
,
y
)
=
{
0.1
if
(
x
=
0
,
y
=
0
)
0.2
if
(
x
=
0
,
y
=
1
)
0.3
if
(
x
=
1
,
y
=
0
)
0.4
if
(
x
=
1
,
y
=
1
)
p(x, y) = \begin{cases} 0.1 & \text{if } (x=0, y=0) \\ 0.2 & \text{if } (x=0, y=1) \\ 0.3 & \text{if } (x=1, y=0) \\ 0.4 & \text{if } (x=1, y=1) \end{cases}
p(x,y)=⎩
⎨
⎧0.10.20.30.4if (x=0,y=0)if (x=0,y=1)if (x=1,y=0)if (x=1,y=1)
计算条件概率 p ( y ∣ x ) p(y|x) p(y∣x)
-
计算边缘概率 p ( x ) p(x) p(x):
p ( x = 0 ) = p ( 0 , 0 ) + p ( 0 , 1 ) = 0.1 + 0.2 = 0.3 p ( x = 1 ) = p ( 1 , 0 ) + p ( 1 , 1 ) = 0.3 + 0.4 = 0.7 p(x=0) = p(0,0) + p(0,1) = 0.1 + 0.2 = 0.3 \\ p(x=1) = p(1,0) + p(1,1) = 0.3 + 0.4 = 0.7 p(x=0)=p(0,0)+p(0,1)=0.1+0.2=0.3p(x=1)=p(1,0)+p(1,1)=0.3+0.4=0.7 -
计算 p ( y ∣ x ) p(y|x) p(y∣x):
p ( y = 0 ∣ x = 0 ) = p ( 0 , 0 ) p ( x = 0 ) = 0.1 0.3 = 1 3 p ( y = 1 ∣ x = 0 ) = p ( 0 , 1 ) p ( x = 0 ) = 0.2 0.3 = 2 3 p ( y = 0 ∣ x = 1 ) = p ( 1 , 0 ) p ( x = 1 ) = 0.3 0.7 ≈ 0.429 p ( y = 1 ∣ x = 1 ) = p ( 1 , 1 ) p ( x = 1 ) = 0.4 0.7 ≈ 0.571 p(y=0|x=0) = \frac{p(0,0)}{p(x=0)} = \frac{0.1}{0.3} = \frac{1}{3} \\ p(y=1|x=0) = \frac{p(0,1)}{p(x=0)} = \frac{0.2}{0.3} = \frac{2}{3} \\ p(y=0|x=1) = \frac{p(1,0)}{p(x=1)} = \frac{0.3}{0.7} \approx 0.429 \\ p(y=1|x=1) = \frac{p(1,1)}{p(x=1)} = \frac{0.4}{0.7} \approx 0.571 p(y=0∣x=0)=p(x=0)p(0,0)=0.30.1=31p(y=1∣x=0)=p(x=0)p(0,1)=0.30.2=32p(y=0∣x=1)=p(x=1)p(1,0)=0.70.3≈0.429p(y=1∣x=1)=p(x=1)p(1,1)=0.70.4≈0.571
计算条件概率 p ( x ∣ y ) p(x|y) p(x∣y)
-
计算边缘概率 p ( y ) p(y) p(y):
p ( y = 0 ) = p ( 0 , 0 ) + p ( 1 , 0 ) = 0.1 + 0.3 = 0.4 p ( y = 1 ) = p ( 0 , 1 ) + p ( 1 , 1 ) = 0.2 + 0.4 = 0.6 p(y=0) = p(0,0) + p(1,0) = 0.1 + 0.3 = 0.4 \\ p(y=1) = p(0,1) + p(1,1) = 0.2 + 0.4 = 0.6 p(y=0)=p(0,0)+p(1,0)=0.1+0.3=0.4p(y=1)=p(0,1)+p(1,1)=0.2+0.4=0.6 -
计算 p ( x ∣ y ) p(x|y) p(x∣y):
p ( x = 0 ∣ y = 0 ) = p ( 0 , 0 ) p ( y = 0 ) = 0.1 0.4 = 0.25 p ( x = 1 ∣ y = 0 ) = p ( 1 , 0 ) p ( y = 0 ) = 0.3 0.4 = 0.75 p ( x = 0 ∣ y = 1 ) = p ( 0 , 1 ) p ( y = 1 ) = 0.2 0.6 = 1 3 p ( x = 1 ∣ y = 1 ) = p ( 1 , 1 ) p ( y = 1 ) = 0.4 0.6 = 2 3 p(x=0|y=0) = \frac{p(0,0)}{p(y=0)} = \frac{0.1}{0.4} = 0.25 \\ p(x=1|y=0) = \frac{p(1,0)}{p(y=0)} = \frac{0.3}{0.4} = 0.75 \\ p(x=0|y=1) = \frac{p(0,1)}{p(y=1)} = \frac{0.2}{0.6} = \frac{1}{3} \\ p(x=1|y=1) = \frac{p(1,1)}{p(y=1)} = \frac{0.4}{0.6} = \frac{2}{3} p(x=0∣y=0)=p(y=0)p(0,0)=0.40.1=0.25p(x=1∣y=0)=p(y=0)p(1,0)=0.40.3=0.75p(x=0∣y=1)=p(y=1)p(0,1)=0.60.2=31p(x=1∣y=1)=p(y=1)p(1,1)=0.60.4=32
总结
从联合概率 p ( x , y ) p(x, y) p(x,y) 得到条件概率 p ( y ∣ x ) p(y|x) p(y∣x) 和 p ( x ∣ y ) p(x|y) p(x∣y) 的关键步骤包括:
- 计算边缘概率 p ( x ) p(x) p(x) 和 p ( y ) p(y) p(y)。
- 使用条件概率的定义,通过联合概率和边缘概率计算条件概率。
这些步骤使得我们能够从联合概率中提取出具体的条件概率,进而实现数据生成和其他任务。需要注意的是,条件概率的计算前提是边际概率不为0。