Neural Networks QUIZ (MCQ QUESTIONS AND ANSWERS)

Principal Component Analysis (PCA)

Gradient Descent

Backpropagation

Weight Initialization

Epoch

Iteration

Batch

Step

Data augmentation

Feature extraction

One-hot encoding

Batch normalization

To set initial values for weights and biases

To update weights during training

To introduce non-linearity

To compute the gradient

Input layer

Hidden layer

Output layer

Support layer

To apply a local operation on the input data

To introduce non-linearity

To update weights

To reduce the dimensionality of the input data

Feature extraction

Feature scaling

One-hot encoding

Data augmentation

Bagging

Regularization

Feature scaling

Weight initialization

To measure the difference between two probability distributions

To introduce non-linearity

To update weights

To compute the gradient

To convert input values into probabilities

To introduce non-linearity

To update weights

To reduce the dimensionality of the input data

Overfitting

Regularization

Generalization

Convergence

To store information from previous time steps

To update weights

To reduce the dimensionality of the input data

To introduce non-linearity

To reduce the spatial dimensions of the feature maps

To apply non-linearity

To learn features from input data

To update weights

Using ReLU activation function

Using a smaller learning rate

Using more hidden layers

Using dropout regularization

To speed up training

To prevent overfitting

To introduce non-linearity

To improve generalization

Node

Neuron

Layer

Weight

Regularization

Hyperparameter tuning

Gradient descent

Feature scaling

Learning rate

Number of hidden layers

Activation function

Input data

Shallow Neural Network

Deep Neural Network

Convolutional Neural Network

Recurrent Neural Network

To control the speed of weight updates

To determine the number of layers

To set the activation function

To initialize the weights

Stochastic Gradient Descent (SGD)

Nesterov Accelerated Gradient (NAG)

AdaBoost

Naive Bayes

To select the best activation function

To minimize the loss function

To determine the number of layers

To set the learning rate

To measure the accuracy of the model

To update the weights

To compute the gradient

To measure the difference between predicted output and actual output

LSTMs have a more complex structure

LSTMs are better at handling long-term dependencies

LSTMs are faster to train

LSTMs require more training data

Convolutional Neural Network (CNN)

Recurrent Neural Network (RNN)

Deep Neural Network (DNN)

Feedforward Neural Network (FNN)

Convolutional Neural Network (CNN)

Recurrent Neural Network (RNN)

Deep Neural Network (DNN)

Long Short-Term Memory (LSTM)

Forward propagation

Backpropagation

Gradient descent

Learning rate

Normalization

Dropout

Activation function

Backpropagation

Sigmoid

ReLU

Tanh

Euclidean

To introduce non-linearity into the model

To perform matrix multiplication

To compute loss

To update weights