Deep Neural Network
Neural Network with multiple hidden layers.
Feedforward Neural Network
Forward and Backward Propagations
Activation Functions
Cost Functions
June 22 2024
Lecture Notes
-> forward propagation for layer l
-> backward propagation for layer l
-> hyperparameters
learning rate, #iterations, #hidden layers L, #hidden units, choice of activation function, momentum, minibatch size...
-> parameters
W[i], b[i] ...
June 23 2024
Lecture Notes
Mismatched train/dev/test distribution
Bias vs. Variance

Last updated