pre
为了便于理解使用只含有一个特征的线性回归进行陈述:
假设函数:
h
θ
(
x
(
i
)
)
=
θ
1
x
(
i
)
+
θ
0
h_{\theta}(x^{(i)})=\theta_{1}x^{(i)}+\theta_{0}
hθ(x(i))=θ1x(i)+θ0
使用MSE损失函数
J
(
θ
0
,
θ
1
)
=
1
2
m
∑
i
=
1
m
(
h
θ
(
x
(
i
)
)
−
y
(
i
)
)
2
J_{(\theta_{0},\theta_{1})} =\frac{1}{2m}\sum_{i=1}^m(h_{\theta}(x^{(i)})-y^{(i)})^2
J(θ0,θ1)=2m1∑i=1m(hθ(x(i))−y(i))2
使用MSE +
L
2
L_2
L2 正则化
J
(
θ
0
,
θ
1
)
=
1
2
m
∑
i
=
1
m
(
h
θ
(
x
(
i
)
)
−
y
(
i
)
)
2
+
1
2
m
∣
∣
θ
1
∣
∣
2
J_{(\theta_{0},\theta_{1})} =\frac{1}{2m}\sum_{i=1}^m(h_{\theta}(x^{(i)})-y^{(i)})^2 + \frac{1}{2m}||\theta_1||^2
J(θ0,θ1)=2m1∑i=1m(hθ(x(i))−y(i))2+2m1∣∣θ1∣∣2