MCQs

Disclaimer:

Educational Use Only:

The website links provided are to be used strictly for educational and knowledge-gaining purposes only.

No Misuse:

Do not misuse the links or the information contained therein for any commercial, illegal, or unethical activities.

Respect for Websites and Businesses:

Respect the intellectual property and business interests of the websites and their owners. Do not attempt to alter, copy, or redistribute content without proper authorization.

Acknowledge Value of Knowledge:

Recognize and value the knowledge and resources shared by these websites. Use the information responsibly and ethically.

No Endorsement:

The provision of these links does not imply endorsement of the linked sites, their content, or their products and services. Users are responsible for evaluating the suitability and accuracy of the information.

User Responsibility:

By accessing these links, you agree to use the information at your own risk and to comply with all applicable laws and website terms of use.

Overall MCQs for Reference

Which of the following layers is always present in a Multi-Layer Perceptron (MLP) architecture?
a. Convolutional Layer
b. Input Layer
c. Pooling Layer
d. Embedding Layer
Answer: b. Input Layergeeksforgeeks+2
What is the primary function of the hidden layers in MLP?
a. Store output values
b. Perform intermediate computations
c. Display results
d. Receive raw inputs
Answer: b. Perform intermediate computationscodesignal+1
In an MLP, each neuron in one layer is connected to every neuron in which layer?
a. The same layer
b. Previous layer
c. Next layer
d. Output layer only
Answer: c. Next layermilvus+1
Which activation function is commonly used in hidden layers of MLP?
a. Softmax
b. Sigmoid
c. ReLU
d. Linear
Answer: c. ReLUwikipedia+1
What does MLP stand for in the context of artificial intelligence?
a. Machine Logical Processor
b. Module Layer Perceptron
c. Multi-Layer Perceptron
d. Model Learning Parameters
Answer: c. Multi-Layer Perceptronmoontechnolabs+1
What type of neural network is an MLP?
a. Recurrent Neural Network
b. Deep Belief Network
c. Feedforward Neural Network
d. Convolutional Neural Network
Answer: c. Feedforward Neural Networkcodesignal+2
Which of the following statements best describes the learning process in MLP?
a. Online learning
b. Backpropagation
c. Genetic algorithm
d. Random selection
Answer: b. Backpropagationlset+1
What do weights in an MLP represent?
a. Output of neurons
b. Strength of connections
c. Type of activation function
d. Bias values
Answer: b. Strength of connectionsdatacamp+1
What does forward propagation in MLP refer to?
a. Adjusting weights backward
b. Moving data from input to output layer
c. Shuffling training data
d. Adding regularization
Answer: b. Moving data from input to output layerturing+1
Bias in MLP neurons helps to:
a. Increase learning rate
b. Adjust activation threshold
c. Remove overfitting
d. Combine inputs
Answer: b. Adjust activation thresholdmilvus+1
Which element introduces non-linearity in MLP models?
a. Weights only
b. Loss function
c. Activation function
d. Input data
Answer: c. Activation functionlset+1
An MLP can approximate any:
a. Discrete function
b. Linear function only
c. Continuous function
d. Boolean function only
Answer: c. Continuous functionwikipedia+1
What happens during backpropagation in an MLP?
a. Inputs are normalized
b. Gradients are calculated and weights are updated
c. Model is saved
d. Output layer is removed
Answer: b. Gradients are calculated and weights are updatedturing+1
The output layer of an MLP is responsible for:
a. Changing weights
b. Final prediction
c. Receiving inputs
d. Data preprocessing
Answer: b. Final predictiongeeksforgeeks+1
Which is NOT a characteristic of MLP architecture?
a. Fully connected layers
b. Non-linear activations
c. Convolutional filters
d. Bias in neurons
Answer: c. Convolutional filtersmilvus+1
What is commonly used to train an MLP efficiently?
a. Stochastic Gradient Descent (SGD)
b. Random forest
c. Genetic algorithms
d. Decision tree
Answer: a. Stochastic Gradient Descent (SGD)turing+1
In MLP, the number of hidden layers affects:
a. The number of outputs
b. The learning rate
c. Model’s complexity
d. Input normalization
Answer: c. Model’s complexitygeeksforgeeks+1
What is the main advantage of using multiple hidden layers in an MLP?
a. Faster training
b. Deeper representation learning
c. Fewer parameters
d. No need for activation functions
Answer: b. Deeper representation learningupgrad+1
What is the purpose of a loss function in MLP training?
a. To mutate the architecture
b. To measure prediction error
c. To shuffle input data
d. To initialize weights
Answer: b. To measure prediction errorturing+1
Which activation function is often used in MLP’s output layer for classification tasks?
a. ReLU
b. Softmax
c. Tanh
d. Sigmoid
Answer: b. Softmaxwikipedia+1
What does ‘feedforward’ mean in the context of MLPs?
a. Data flows backward
b. Data flows one way from input to output
c. Data is fed into all layers simultaneously
d. Output is reused as input
Answer: b. Data flows one way from input to outputmoontechnolabs+1
Which of the following is NOT a part of the MLP architecture?
a. Input layer
b. Output layer
c. Pooling layer
d. Hidden layer
Answer: c. Pooling layermilvus+1
The complexity of an MLP is mainly increased by adding:
a. More input features
b. More hidden layers or more neurons per layer
c. More data samples
d. Larger batch sizes
Answer: b. More hidden layers or more neurons per layerupgrad+1
What property allows MLPs to solve non-linear problems?
a. Linear weights
b. Non-linear activation functions
c. Bias units
d. High learning rate
Answer: b. Non-linear activation functionslset+1
In an MLP trained for regression, which activation function is usually preferred in the output layer?
a. Sigmoid
b. Linear
c. Softmax
d. ReLU
Answer: b. Lineargeeksforgeeks+1
What does an epoch represent during an MLP training process?
a. One pass through the entire dataset
b. One pass through a mini-batch
c. One update of a single weight
d. One calculation of output
Answer: a. One pass through the entire datasetscribd
How is overfitting in MLPs usually prevented?
a. Increasing learning rate
b. Adding regularization like dropout
c. Removing output layer
d. Using only linear activation
Answer: b. Adding regularization like dropoutscribd+1
Which term describes the network’s ability to represent complex functions?
a. Linearity
b. Non-linearity
c. Generalization
d. Overfitting
Answer: b. Non-linearitylset+1
What is the main limitation of single-layer perceptrons that MLPs overcome?
a. Memory usage
b. Ability to model linear functions
c. Ability to model non-linear functions
d. Communication speed
Answer: c. Ability to model non-linear functionswikipedia+1
The output values of an MLP are determined by:
a. Weight initialization
b. Forward propagation through the network
c. Data preprocessing methods only
d. The number of epochs
Answer: b. Forward propagation through the networkmoontechnolabs+1
What typically happens as the number of neurons in MLP hidden layers increases?
a. Model learns richer representations
b. Training data decreases
c. Output layer is removed
d. Bias is ignored
Answer: a. Model learns richer representationscodesignal+2
What is a typical use-case of MLPs in artificial intelligence?
a. Image filtering only
b. Classification and regression tasks
c. Only anomaly detection
d. Clustering only
Answer: b. Classification and regression tasksupgrad+1
Which architecture is MLP most closely related to?
a. Recurrent Neural Network
b. Convolutional Neural Network
c. Fully Connected Network
d. Decision Tree
Answer: c. Fully Connected Networkmilvus+1
Which of the following increases the computational cost of an MLP?
a. Batch normalization
b. More layers and neurons
c. Linear activation
d. Small batch size
Answer: b. More layers and neuronsdatacamp+1
Which is a key characteristic of neurons in all MLP layers?
a. Non-linearity through activation functions
b. Fixed, hardcoded outputs
c. Shared weights
d. No bias units allowed
Answer: a. Non-linearity through activation functionslset+1
Which optimizer is frequently used for MLP model training?
a. Adam
b. Lasso
c. K-means
d. DBSCAN
Answer: a. Adamturing+1
What allows the MLP to update its weights during training?
a. Loss function and optimizer through backpropagation
b. Only random initialization of weights
c. Additional input layers
d. Only user intervention
Answer: a. Loss function and optimizer through backpropagationturing+1
How are the inputs combined in each neuron of an MLP?
a. By summing weighted inputs and adding bias
b. By random selection
c. By applying dropout first
d. By clustering them
Answer: a. By summing weighted inputs and adding biasdatacamp+1
The depth of an MLP is determined by:
a. The number of output neurons
b. The number of hidden layers
c. The size of the training dataset
d. The size of input features
Answer: b. The number of hidden layerscodesignal+1
Which problem is specifically solved by using an MLP over a linear model?
a. Modeling non-linear data relationships
b. Solving decision trees faster
c. Reducing dimensionality
d. Improving clustering accuracy
Answer: a. Modeling non-linear data relationshipswikipedia+1
What is the role of an activation function in MLP?
a. Provide regularization
b. Introduce non-linearity
c. Remove bias
d. Limit weight updates
Answer: b. Introduce non-linearitylset+1
In MLP training, weight update occurs after calculation of:
a. The output
b. The gradient of the loss function
c. The bias term
d. Only input normalization
Answer: b. The gradient of the loss functionmilvus+1
What does a neuron compute in an MLP?
a. Weighted sum of inputs, applies activation function
b. Only the product of weights and bias
c. Only the maximum value input
d. Identity function
Answer: a. Weighted sum of inputs, applies activation functiondatacamp+1
Which of the following is NOT used in a typical MLP training process?
a. Forward propagation
b. K-means clustering
c. Loss function
d. Backpropagation
Answer: b. K-means clusteringturing+1
What does dropout regularization do in MLP architecture?
a. Removes neurons during training to prevent overfitting
b. Fixes the output layer
c. Only adjusts weight initialization
d. Calculates output faster
Answer: a. Removes neurons during training to prevent overfittingscribd+1
Which layer’s outputs are used as the final output of the MLP?
a. Input layer
b. All hidden layers
c. The last hidden layer
d. Output layer
Answer: d. Output layermoontechnolabs+1
For increasing model interpretability in MLP, one should:
a. Reduce the number of layers
b. Use linear activation functions
c. Remove biases
d. Increase the number of neurons
Answer: a. Reduce the number of layersupgrad+1
Which is the most common reason for vanishing gradients in deep MLPs?
a. Sigmoid or tanh activation functions
b. ReLU activation function
c. Linear activation function
d. Large batch size
Answer: a. Sigmoid or tanh activation functionswikipedia
In the context of MLP, training stops when:
a. All weights reach zero
b. The loss function converges
c. The bias values are maximized
d. Batch size equals input size
Answer: b. The loss function convergesmilvus+1
What dataset is often used to benchmark image classification performance of an MLP?
a. MNIST
b. ImageNet
c. CIFAR-10
d. COCO
Answer: a. MNISTturing+1

Blog

Systemethic Technologies

Leave a Reply Cancel reply

Useful links

Other Terms & Policies

Contact Info