Bayes基本概念
Prior(先验):
p
(
z
)
p(z)
p(z) 看到数据前的猜测;
Posterior(后验):
p
(
z
∣
x
)
p(z|x)
p(z∣x) 看到数据后的猜测;
Likelihood(似然):
p
(
x
∣
z
)
p(x|z)
p(x∣z) 也是条件概率,在
z
z
z假设下数据
x
x
x的概率;
Evidence(证据):
p
(
x
)
p(x)
p(x)即数据本身;
注:
x
x
x 是数据,
z
z
z 是变量
p ( z ∣ x ) = p ( x ∣ z ) ∗ p ( z ) p ( x ) p(z|x)=\frac{p(x|z) * p(z)} {p(x)} p(z∣x)=p(x)p(x∣z)∗p(z)
根据条件概率:
p
(
z
∣
x
)
=
p
(
x
,
z
)
∫
p
(
x
,
z
)
d
z
=
p
(
x
,
z
)
p
(
x
)
p(z|x)=\frac{p(x,z)} {\int p(x,z)dz}=\frac{p(x,z)} {p(x)}
p(z∣x)=∫p(x,z)dzp(x,z)=p(x)p(x,z)