NEW
Font size
WorksheetsArtificial Neural Networks (Part 1- Part 3) Quiz
Total questions: 20
Worksheet time: 10mins
What does the input layer of an ANN contain?
Neurons
Bias values
Input variables
Transfer functions
Which layer in ANN makes the final prediction?
Input layer
Hidden layer
Output layer
Training layer
What is the purpose of a hidden layer in ANN?
Collect raw input
Make final decisions
Perform internal processing
Display results
Which of the following is NOT part of a neuron?
Weight
Bias
Input variable
Transfer function
Which formula correctly represents net input (n) for a neuron with a single input?
n = w - b
n = w + b
n = wp + b
n = p - b
What happens to input data in the hidden layer?
It's ignored
It's passed directly to the output
It's multiplied by weights and added to biases
It's stored as memory
Which of the following statements about neurons with multiple inputs is TRUE?
They must have equal weights
They always produce binary outputs
Their net input is calculated using vector multiplication
They can't be used in ANN
Which transfer function produces only binary output (0 or 1)?
Linear
ReLU
Hard-limiter
Tanh
Which activation function is commonly used in the output layer for regression problems?
Tanh
ReLU
Sigmoid
Linear
Which activation function helps prevent the vanishing gradient problem?
Sigmoid
Tanh
ReLU
Hard-limiter
What is a disadvantage of using sigmoid function in hidden layers?
Not continuous
Vanishing gradient problem
Only works with text
Cannot be differentiated
What is the range of output for the tanh activation function?
0 to 1
-1 to 1
-∞ to ∞
0 to ∞
Which function is most suitable for output layer in binary classification tasks?
ReLU
Tanh
Sigmoid
Linear
Which layer typically shares the same transfer function in an ANN?
Input layer
Hidden layers
Output layer
All layers
What is the main reason ReLU is preferred in hidden layers over sigmoid?
Simpler code
Better output precision
Less vanishing gradient issue
Better binary prediction
An MLP network must contain at least:
One input and one output layer only
One hidden layer only
One input, one hidden, and one output layer
Only two layers
MLP uses which type of learning method?
Unsupervised
Reinforcement
Supervised
Evolutionary
Which transfer functions are commonly used in hidden layers of MLP?
ReLU, Sigmoid, Tanh
Linear only
Hard-limiter and Linear
Linear and Tanh only
Which activation function is not used in hidden layers in modern MLP?
ReLU
Sigmoid
Tanh
Hard-limiter
Which line of Python code adds an output layer with linear activation in Keras?
model.add(Dense(1, activation='relu'))
model.add(Dense(1, activation='sigmoid'))
model.add(Dense(1, activation='linear'))
model.add(Dense(1, activation='tanh'))
