ATI quiz

Which of the following activation functions can lead to vanishing gradients?

• ReLU

• Tanh

• 1.  Leaky ReLU

• 1.  None of the above

Which of the following is true about Batchnorm?

• Batchnorm is another way of performing dropout

• Batchnorm makes training faster

• In Batchnorm, the mean is computed over the features

• Batchnorm is a non-linear transformation to center the dataset around the origin

When should multi-task learning be used?

• When your problem involves more than one class label

• When two tasks have the same dataset

• When you have a small amount of data for a particular task that would benefit from the large dataset of another task

• When the tasks have datasets of different formats (text and images).

What is Error Analysis?

• The process of analyzing the performance of a model through metrics such as precision, recall or F1-score.

• The process of scanning mis-classified examples to identify weaknesses of a model

• The process of tuning hyperparameters to reduce the loss function during training

•  The process of identifying which parts of your model contributed to the error

Which of the following is a non-iterative method to generate adversarial examples?

• 1.  Non-Saturating Cost Method

• Input Optimization Method

• Adversarial Training

• Logit Pairing

• 1.  Fast Gradient Sign Method

Which, if any, of the following propositions is true about fully-connected neural networks (FCNN)?

• A FCNN with only linear activations is a linear network.

• In a FCNN, there are connections between neurons of a same layer.

• In a FCNN, the most common weight initialization scheme is the Zero initialization, because it leads to faster and more robust training

• None of the above

How many layers Deep learning algorithms are constructed?

• 2

• 3

• 4

• 5

In CNN, having max pooling always decrease the parameters?

• True

• False

• Can be true and false

• Can not say

Which of the following is NOT the CNN layers?

• Convolutional Layer

• Pooling Layer

• Code Layer

• Fully connected Layer

Which of the following is/are Common uses of RNNs?

• BusinessesHelp securities traders to generate analytic reports

• Detect fraudulent credit-card transaction

• Provide a caption for images

• All of the above