Boost Your Neural Networks: Quiz on ReLU, Sigmoid, and Tanh Activation Functions — Questions & Answers

This quiz contains 10 questions. Below is a complete reference of all questions, answer choices, and correct answers. You can use this section to review after taking the interactive quiz above.

1. Question 1: Basic Concept

   Which of the following activation functions outputs values strictly between 0 and 1 for any real input?

   • ReLU
   • Sigmoid
   • Tanh
   • Softplus
   • Swish
   Show correct answer

   Correct answer: Sigmoid

2. Question 2: Function Form

   What is the mathematical expression for the ReLU activation function?

   • max(0, x)
   • 1 / (1 + e^-x)
   • tanh(x)
   • min(1, x)
   • log(1+e^x)
   Show correct answer

   Correct answer: max(0, x)

3. Question 3: Output Range

   If you input large negative values into the Tanh activation function, what is the approximate output?

   • Close to 1
   • Close to 0
   • Close to -1
   • Exactly 0
   • Exactly 1
   Show correct answer

   Correct answer: Close to -1

4. Question 4: Identity Comparison

   Which activation function among ReLU, Sigmoid, and Tanh is non-linear but does not squash negative values to positive outputs?

   • Sigmoid
   • ReLU
   • Tanh
   • Softmax
   • Leaky ReLU
   Show correct answer

   Correct answer: ReLU

5. Question 5: Derivative Behavior

   For which input does the derivative of the Sigmoid activation function reach its maximum value?

   • x = 0
   • x = 1
   • x = -1
   • x = infinity
   • x = -infinity
   Show correct answer

   Correct answer: x = 0

6. Question 6: Scenario Application

   If you want your activation function's output to cover both negative and positive ranges symmetrically, which should you use?

   • Sigmoid
   • Tanh
   • ReLU
   • Softplus
   • Linear
   Show correct answer

   Correct answer: Tanh

7. Question 7: Zero Value Handling

   What value does ReLU return if the input is zero?

   • Zero
   • One
   • Negative one
   • Infinity
   • Half
   Show correct answer

   Correct answer: Zero

8. Question 8: Activation Function Saturation

   Which activation function is prone to the vanishing gradient problem because its output saturates for large positive or negative inputs?

   • Tanh
   • ReLU
   • Leaky ReLU
   • Swish
   • Poly
   Show correct answer

   Correct answer: Tanh

9. Question 9: Probabilistic Output

   If you need an output interpreted as a probability, which activation function is most suitable at an output layer?

   • ReLU
   • Tanh
   • Sigmoid
   • Piecewise Linear
   • Exponential
   Show correct answer

   Correct answer: Sigmoid

10. Question 10: Negative Inputs

    When using the ReLU activation function, what is the output for an input value of -5?

    • -5
    • 0
    • 5
    • 1
    • -1