Appendix
Chapter 1, Artificial Neural Network Fundamentals
- What are the various layers in a neural network?
Input, hidden, and output.
- What is the output of feedforward propagation?
Predictions that help in calculating the loss value.
- How is the loss function of a continuous dependent variable different from that of a binary dependent variable or a categorical dependent variable?
Mean squared error (MSE) is the generally used loss function for continuous dependent variables, and binary cross-entropy is generally used for binary dependent variables. Categorical cross-entropy is used for categorical dependent variables.
- What is stochastic gradient descent?
It is the process of reducing loss by adjusting weights in the direction of decreasing gradient.
- What does a backpropagation exercise do?
It computes the gradients of all weights with respect to loss using the chain rule.
-
...
