1. hypothsis

 

2. cost function: 

3. Goal: 

4. Gradient descent algorithm

repeat until convergence {

　　            

　　(for j = 0 and j = 1)

}

note: simultaneous update

α:learning rate

if α is too small, gradient descent can be slow.

if α is too large, gradient descent can overshoot the minimum. It may fail to converge, or even diverge.

5. Gradient descent algorithm for one variable

repeat until convergence {

　　

　　

}

6. "batch" gradient descent: each step of gradient descent uses all the training examples