You now know about individual neurons, activation functions, and how to stack them into layers. Let's practice thinking about the overall structure, or architecture, of a neural network.
Engagement Message
Ready to design some networks?
Type
Sort Into Boxes
Practice Question
Sort these terms into their correct category.
Labels
- First Box Label: Part of a Neuron
- Second Box Label: Part of a Network
First Box Items
- Weight
- Bias
- Activation
Second Box Items
- Input Layer
- Hidden Layer
- Output Layer
Type
Multiple Choice
Practice Question
A neural network is designed to predict house prices (a single continuous number). How many neurons should its output layer have?
A. 1 B. 10 C. 100 D. It depends on the number of inputs.
Suggested Answers
- A - Correct
- B
- C
- D
Type
Fill In The Blanks
Markdown With Blanks
Fill in the blanks to describe the flow of information in a network.
Data first enters the [[blank:input layer]]. It is then processed by one or more [[blank:hidden]] layers, which extract patterns. Finally, the [[blank:output layer]] produces the network's final prediction.
Suggested Answers
- input layer
- hidden
- output layer
Type
Multiple Choice
Practice Question
A network has an input layer that accepts 50 features. The first hidden layer has 25 neurons, and the second hidden layer has 10 neurons. How many inputs does each neuron in the second hidden layer receive?
A. 50 B. 25 C. 10 D. 1
Suggested Answers
- A
- B - Correct
- C
- D
Type
Swipe Left or Right
Practice Question
Swipe to match the layer with its primary role in the network.
Labels
- Left Label: Feature Extraction
- Right Label: Final Decision
Left Label Items
- First hidden layer
- Middle hidden layers
- Learns simple patterns
- Combines simple patterns
Right Label Items
- Output layer
- Produces the prediction
