Deep Learning - Chorale Prelude :

Q.1 __________ is a Neural Nets way of classifying inputs.

Ans : Learning

Q.2 Recurrent Neural Networks are best suited for Text Processing.

Ans : True

Q.3 Neural Networks Algorithms are inspired from the structure and functioning of the Human Biological Neuron.

Ans : True

Q.4 The rate at which cost changes with respect to weight or bias is called _________.

Ans : Gradient

Q.5 __________ is a recommended Model for Pattern Recognition in Unlabeled Data.

Ans : Autoencoders

Q.6 What does LSTM stand for?

Ans : Long Short Term Memory

Q.7 RELU stands for __________.

Ans : Rectified Linear Unit

Q.8 All the Visible Layers in a Restricted Boltzmannn Machine are connected to each other.

Ans : False

Q.9 A Deep Belief Network is a stack of Restricted Boltzmann Machines.

Ans : True

Q.10 All the neurons in a convolution layer have different Weights and Biases.

Ans : False

Q.11 A Shallow Neural Network has only one hidden layer between Input and Output layers.

Ans : True

Q.12 GPU stands for __________.

Ans : Graphics Processing Unit

Q.13 Autoencoders cannot be used for Dimensionality Reduction.

Ans : True

Q.14 A _______________ matches or surpasses the output of an individual neuron to a visual stimuli.

Ans : Convolution

Q.15 Name the component of a Neural Network where the true value of the input is not observed.

Ans : Activation Function

Q.16 What is the best Neural Network Model for Temporal Data?

Ans : Recurrent Neural Network

Q.17 The measure of Difference between two probability distributions is know as ___________.

Ans : KL Divergence

Q.18 All the Visible Layers in a Restricted Boltzmann Machine are connected to each other.

Ans : True

Q.19 In a Neural Network, all the edges and nodes have the same Weight and Bias values.

Ans : True

Q.20 What is the difference between the actual output and generated output known as?

Ans : Output Difference

Q.21 Process of improving the accuracy of a Neural Network is called __________.

Ans : Training

Q.22 Autoencoders are trained using _________.

Ans : Back Propagation

Q.23 Data Collected from Survey results is an example of __________.

Ans : Structured Data

Q.24 Support Vector Machines, Naive Bayes and Logistic Regression are used for solving ___________ problems.

Ans : Clustering

Q.25 Recurrent Network can input Sequence of Data Points and Produce a Sequence of Output.

Ans : True

Q.26 Gradient at a given layer is the product of all gradients at the previous layers.

Ans : True

Q.27 What is the method to overcome the Decay of Information through time in RNN known as?

Ans : Back Propagation

Q.28 Why is the Pooling Layer used in a Convolution Neural Network?

Ans : Dimension Reduction

Q.29 Prediction Accuracy of a Neural Network depends on _______________ and ______________.

Ans : Input and Output

Q.30 ____________ works best for Image Data.

Ans : AutoEncoders

Q.31 ____________ models are best suited for Recursive Data.

Ans : Recursive Neural Networks