WorksheetsANN_Unit 1
Total questions: 20
Worksheet time: 6mins
Which of the following best describes an Artificial Neural Network (ANN)?
A set of predefined rules
A system inspired by the human brain for processing information
A database management technique
A programming language
What was a significant limitation of early Artificial Neural Networks?
They were only used for image processing
They were computationally expensive
They could not learn from experience
They required large amounts of data
What is a primary function of biological neurons that is not directly replicated in artificial neurons?
Signal transmission through synapses
Activation functions
Layered architecture
Weight adjustment
How do artificial neurons typically process information compared to biological neurons?
Through complex biochemical reactions
By generating electrical impulses
Using mathematical functions
Through chemical signals
Which of the following statements is true regarding the learning process of biological neurons versus artificial neurons?
Artificial neurons require labeled data for learning
Biological neurons learn through reinforcement only
Artificial neurons can learn without any data
Biological neurons do not adapt over time
What is the primary purpose of a single perceptron model in machine learning?
To generate synthetic data
To classify data into two categories
To perform regression analysis
To classify data into multiple categories
Which of the following components is essential for the functioning of a single perceptron?
Activation function
Convolutional layers
Pooling layers
Recurrent connections
In a single perceptron model, what does the weight associated with an input signify?
The number of inputs to the perceptron
The importance of the input in the decision-making process
The type of activation function used
The output of the perceptron
What is the role of backpropagation in training neural networks?
To initialize the weights of the network
To update the weights based on the error gradient
To generate new training data
To evaluate the performance of the model
What is the output of a single perceptron when the weighted sum of inputs exceeds the threshold?
It cannot be determined
It depends on the activation function
1
0
In a simple perceptron model, what is the effect of increasing the weights of the inputs?
It decreases the output
It has no effect on the output
It increases the output if the weighted sum exceeds the threshold
It makes the perceptron more complex
Which of the following best describes the activation function in a simple perceptron?
A logarithmic function
A polynomial function
A step function
A linear function
What is the output of a step activation function when the input is below the threshold?
It cannot be determined
1
It varies based on the input
0
Which of the following is a characteristic of the step activation function?
It is used primarily in regression tasks
It is continuous and differentiable
It produces a binary output
It can output multiple values
In which scenario is a step activation function most commonly used?
In reinforcement learning
In multi-class classification problems
In binary classification problems
In unsupervised learning tasks
What effect does increasing the bias in a perceptron have on the output?
It shifts the activation function to the left
It has no effect on the output
It increases the threshold for activation
It decreases the output regardless of the inputs
In the context of neural networks, what is the primary purpose of weights?
To define the architecture of the network
To represent the strength of the connection between neurons
To adjust the learning rate
To determine the output of the network
How does the bias term in a neural network influence the model's performance?
It allows the model to fit the data better by shifting the activation function
It has no impact on the model's performance
It reduces the complexity of the model
It only affects the output when the inputs are zero
What happens during the backpropagation phase of training in a neural network?
Inputs are normalized before training
Errors are propagated backward to update weights
Output is generated without learning
Neurons are randomly initialized
In a single-layer perceptron, how are input features combined to produce the output?
Through a series of hidden layers
By applying an activation function directly to the inputs
By computing the dot product of input features and weights
By using a softmax function
