1. Normalizing inputs
Slide 1

Slide 2

Slide 3

2. Vanishing / Exploding
Slide 1

Slide 2

3. Weight Initialization for Deep Networks
Slide 1

Slide 2

4. Numerical approximation of gradients
Slide 1

Slide 2

Slide 3

5. Gradient checking
Slide 1

Slide 2

Slide 3

6. Gradient Checking Implementation Notes
Slide 1

Slide 2

网友评论