Wiki: CS231n Lecture 7 – Training Neural Networks, part II


Update rules, ensembles, data augmentation, transfer learning

Lecture Slides
Lecture Video
Neural Nets notes 3

Discussion points:

  • How does Nesterov Momentum allow for look ahead?
  • How are adaptive learning rate methods like Adam helpful?

Hands on coding:
Replicate this python notebook and try to improve the model accuracy