logo

Online Exam Quiz

questionQuestion  1 to 30

Exit Quiz

Which of the following techniques is used for regularization in deep learning?

  • Dropout
  • Batch Normalization
  • L1 Regularization
  • All of the above
Mark for review

Which type of layer is used to increase the non-linearity of a neural network?

  • Fully Connected Layer
  • Convolutional Layer
  • Pooling Layer
  • Activation Layer
Mark for review

Which technique is used to normalize the input to a neural network, leading to faster training and better generalization?

  • Dropout
  • Batch Normalization
  • L2 Regularization
  • Weight Initialization
Mark for review

Which of the following techniques is used to reduce the risk of overfitting in deep learning?

  • L1 Regularization
  • L2 Regularization
  • Dropout
  • All of the above
Mark for review

Which of the following techniques is used to prevent overfitting in deep learning?

  • Data augmentation
  • Dropout
  • Early stopping
  • All of the above
Mark for review

What is the primary purpose of activation functions in deep learning?

  • Increase model complexity
  • Speed up model training
  • Introduce non-linearity
  • Regularize the model
Mark for review

Which of the following techniques is used to address the problem of classifying imbalanced datasets in deep learning?

  • Data augmentation
  • Oversampling
  • Weighted loss functions
  • All of the above
Mark for review

Which technique is commonly used for transfer learning in deep learning?

  • Fine-tuning
  • Regularization
  • Dropout
  • Gradient clipping
Mark for review

In deep learning, what does the term "backpropagation" refer to?

  • The process of passing data through the network
  • The process of adjusting model parameters based on the error
  • The process of optimizing hyperparameters
  • The process of selecting the best architecture for a problem
Mark for review

Which type of layer is typically used to connect different parts of a neural network, allowing gradients to flow during backpropagation?

  • Fully Connected Layer
  • Convolutional Layer
  • Recurrent Layer
  • Skip Connection
Mark for review

What is the purpose of the learning rate in gradient descent optimization algorithms?

  • To determine the size of the weight updates
  • To control the number of epochs
  • To decide the number of layers in the network
  • To specify the size of the training dataset
Mark for review

What is the purpose of data augmentation in deep learning?

  • Increase model capacity
  • Reduce model complexity
  • Increase training data diversity
  • Decrease training time
Mark for review

Which technique is used to reduce the dimensionality of the input data in deep learning?

  • Feature scaling
  • Principal Component Analysis (PCA)
  • Dropout
  • Batch Normalization
Mark for review

Which of the following techniques is used to visualize the features learned by a neural network?

  • Principal Component Analysis (PCA)
  • t-Distributed Stochastic Neighbor Embedding (t-SNE)
  • Gradient Descent
  • Ridge Regression
Mark for review

Which of the following is a drawback of using unsupervised pre-training in deep learning?

  • Requires more labeled data
  • Increases computational cost
  • Leads to overfitting
  • None of the above
Mark for review

Which of the following activation functions is generally not used in hidden layers due to vanishing gradient problems?

  • ReLU
  • Tanh
  • Sigmoid
  • Leaky ReLU
Mark for review

In deep learning, what is the purpose of the "padding" parameter in convolutional neural networks?

  • To control the size of the filter
  • To specify the number of filters in each layer
  • To add zeros around the input data to maintain spatial dimensions
  • To control the stride of the filter
Mark for review

Which technique is used to address the problem of exploding gradients during training?

  • Gradient clipping
  • Learning rate decay
  • Batch normalization
  • Xavier initialization
Mark for review

What is the purpose of the learning rate scheduler in deep learning?

  • To adjust the learning rate during training based on the validation performance
  • To control the number of epochs in training
  • To initialize the weights of the neural network
  • To normalize the input data
Mark for review

Which type of neural network architecture is used for anomaly detection?

  • Convolutional Neural Network (CNN)
  • Long Short-Term Memory Network (LSTM)
  • Autoencoder
  • Recurrent Neural Network (RNN)
Mark for review

Which optimization algorithm is commonly used to train deep neural networks?

  • Gradient Descent
  • Stochastic Gradient Descent (SGD)
  • Adam
  • All of the above
Mark for review

Which of the following techniques is used to reduce the computational cost of training deep neural networks?

  • Gradient Descent
  • Stochastic Gradient Descent
  • Mini-batch Gradient Descent
  • All of the above
Mark for review

Which of the following is not a common problem encountered in training deep neural networks?

  • Vanishing gradients
  • Exploding gradients
  • Underfitting
  • Overfitting
Mark for review

Which deep learning architecture is commonly used for image classification tasks?

  • Recurrent Neural Networks (RNNs)
  • Convolutional Neural Networks (CNNs)
  • Long Short-Term Memory Networks (LSTMs)
  • Autoencoders
Mark for review

Which of the following is a pre-trained deep learning model commonly used for various tasks?

  • AlexNet
  • VGG
  • ResNet
  • All of the above
Mark for review

What is the purpose of the softmax function in deep learning?

  • Introduce non-linearity
  • Normalize the output probabilities
  • Prevent overfitting
  • Reduce computational complexity
Mark for review

Which layer in a convolutional neural network is responsible for reducing the spatial dimensions of the input?

  • Convolutional Layer
  • Activation Layer
  • Pooling Layer
  • Fully Connected Layer
Mark for review

Which type of neural network architecture is commonly used for natural language processing tasks?

  • Convolutional Neural Networks (CNNs)
  • Long Short-Term Memory Networks (LSTMs)
  • Recurrent Neural Networks (RNNs)
  • Autoencoders
Mark for review

Which activation function is preferred in the output layer of a multi-class classification problem?

  • ReLU
  • Sigmoid
  • Tanh
  • Softmax
Mark for review

Which technique is commonly used to handle missing data in deep learning?

  • Mean imputation
  • Median imputation
  • Median imputation
  • None of the above
Mark for review

blinkerOnline

  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
  • 28
  • 29
  • 30
  • Answered
  • Not Answered
  • Marked for Review

Get a 1:1 Mentorship call with our Career Advisor

Book free session