机器学习第二课

梯度下降算法:
Repeat
{
${\theta }_j={\theta }_j-\alpha \frac{\partial }{\partial {\theta }_j}J({\theta }_0,{\theta }_1...{\theta }_n)$
}simultaneously update for every j=0,1…n)

$${\theta }_j={\theta }_j-\alpha \frac{1}{m}\sum _{i=1}^m(h_{\theta }(x^{(i)})-y^{(i)})x_j^{(i)}$$
Feature Scaling以及Mean normalizaition

$\alpha$太大：slow convergence
$\alpha$太小：J($\theta$) mat not decrease on every iteration,may not converge
尝试不同的$\alpha$，绘制J($\theta$)随迭代次数变化的曲线

polynominal regression(多项式回归)

Normal equation(正规方程)

$\frac{\partial }{\partial {\theta }_j}J(\theta )=0$ for every j

Gradient Descent 和 Normal Equation各自的优缺点