Building Neural Networks
Neural Network Basics
Layers and neurons
A neural network is made of layers. Each layer transforms its input into a new representation. The simplest networks have an input layer, one or more hidden layers, and an output layer.
Activation functions
Activation functions add nonlinearity, allowing networks to learn complex patterns. Common examples include ReLU and sigmoid. Without nonlinear activations, many networks would behave like a single linear model.
Network Terms
Beginner Rule
Start with small networks. Simpler models are easier to debug and often work surprisingly well.
Why do neural networks use activation functions?
Activations let networks model complex relationships.
Correct answer: To add nonlinearity
What is the role of the output layer?
The output layer maps internal features to the answer.
Correct answer: It produces the final prediction.
Which layer processes intermediate features?
Hidden layers transform data between input and output.
Correct answer: Hidden layer