The vast bulk of data is unstructured and without labels. Shallow neural networks struggle to recognise important structures in, say, audio, visual, or textual data. This kind of data contains hidden structures that can be found by deep networks. You’ll use the TensorFlow library in this course to apply deep learning to various types of data in order to address real-world issues.

### Building Deep Learning Models with TensorFlow Week 01 Quiz Answers

#### Quiz : Deep Neural Networks and TensorFlow

Q1. Which statement is FALSE about TensorFlow?

- TensorFlow is an open source library.
**TensorFlow library is not proper for handling Machine Learning Problems.**- TensorFlow has a C/C++ backend as well as Python modules.
- TensorFlow is well suited for handling Deep Learning Problems.
- All of the above.

Q2. What is/are the main reason(s) of the increasing popularity of Deep Learning?

**The dramatic increases in computer processing capabilities.**- The availability of massive amounts of data for training computer systems.
- The advances in machine learning algorithms and research.
- All of the above

Q3. What is a Data Flow Graph?

- A flowchart describing an algorithm.
- A cartesian (x, y) chart.
- A graphics user interface.
**A representation of data dependencies between operations.**- None of the above

Q4. Which statement is TRUE about TensorFlow?

- Runs on GPU only.
- Runs on CPU only.
**Runs on CPU and GPU.**

Q5. Why is TensorFlow the proper library for Deep Learning?

- It has extensive built-in support for deep learning.
- It provides a collection of trainable mathematical functions that are useful for neural networks.
- It will benefit from TensorFlow’s auto-differentiation and suite of first-rate optimizers.
**All of the above.**

### Building Deep Learning Models with TensorFlow Week 02 Quiz Answers

#### Quiz : Convolutional Neural Networks

Q1. What can be achieved with **convolution** operations on Images?

- Edge Detection
- Image Smoothing
- Image Blurring
- Noise Filtering
**All of the above**

Q2. For convolution, it is better to store images in a TensorFlow Graph as

- Variable
- CSV file
- Numpy array
**Placeholder**- None of the above

Q3. Which of the following statements is TRUE about Convolution Neural Networks (CNNs)?

- CNN can be applied ONLY on Image data
**CNN can be applied on ANY 2D and 3D array of data.**- CNN can be applied ONLY on Text and Speech data.
- CNN can be applied ONLY on Image and Text data.
- All of the above

Q4. Which of the following Layers can be part of Convolution Neural Networks (CNNs)

- Relu
- Softmax
- Maxpooling
- Dropout
**All of the above**

Q5. The objective of the Activation Function is to:

- Reduce the Size of the Network
**Handle Non-Linearity in the Network**- Handle Linearity in the Network
- Increase the Size of the Network
- None of the above

### Building Deep Learning Models with TensorFlow Week 03 Quiz Answers

#### Quiz : Recurrent Neural Networks

Q1. What is a Recurrent Neural Network?

**A Neural Network that can recur to itself, and is proper for handling sequential data**- An infinite layered Neural Network which is proper for handling structured data
- A special kind of Neural Network to predict weather
- A Markovian model to handle temporal data

Q2. What is NOT TRUE about RNNs?

**RNNs are VERY suitable for sequential data.**- RNNs need to keep track of states, which is computationally expensive.
- RNNs are very robust against vanishing gradient problem.

Q3. What application(s) is(are) suitable for RNNs?

- Speech Recognition
- Natural Language Processing
- Video context retriever
- Estimating temperatures from weather data
**All of the above**

Q4. Why are RNNs susceptible to issues with their gradients?

- Gradients can grow exponentially
- Gradients can quickly drop and stabilize at near zero
- Propagation of errors due to the recurrent characteristic
- Numerical computation of gradients can drive into instabilities
**All of the above**

Q5. What is TRUE about LSTM gates?

- The Read Gate in LSTM, is responsible for writing data into the memory cell.
- The Write Gate in LSTM, reads data from the memory cell and sends that data back to the network.
**The Forget Gate, in LSTM maintains or deletes data from the information cell.**- The Read Gate in LSTM, determine how much old information to forget

### Building Deep Learning Models with TensorFlow Week 04 Quiz Answers

#### Quiz : Restricted Boltzmann Machines

Q1. What is the main application of RBM?

- Collaborative filtering
- Feature extraction
- Data dimensionality reduction
**All of the above**

Q2. How many layers does an RBM (Restricted Boltzmann Machine) have?

- 3
- 4
**2**- Infinte
- All of the above

Q3. How does an RBM compare to a PCA?

**Both can regenerate input data**- PCA cannot generate original data
- PCA is another type of Neural Network
- RBM cannot reduce dimensionality
- All of the above

Q4. Which statement is TRUE about RBM?

- At the hidden layer’s nodes, X is multiplied by a W (weight matrix) and added to h_bias.
- Each node in the first layer has a bias.
- The RBM reconstructs data by making several forward and backward passes between the visible and hidden layers.
- It is a Boltzmann machine, but with no connections between nodes in the same layer.
**All of the above**

Q5. Which statement is TRUE statement about an RBM?

- The Positive phase of an RBM increases the probability of training data.
- The Negative phase of an RBM decreases the probability of samples generated by the model.
- Contrastive Divergence (CD) is used to approximate the negative phase of an RBM.
- The objective function is to maximize the likelihood of our data being drawn from the reconstructed data distribution.
**All of the above**

### Building Deep Learning Models with TensorFlow Week 05 Quiz Answers

#### Quiz : Autoencoders

Q1. What are Autoencoders?

- A Neural Network where different layer inputs are controlled by gates
- A Neural Network that is trained to attempt to copy its input to its output
- A Neural Network that learns all the weights by using labeled data
- A Neural Network that is designed to replace Non-Linear Regression
**All of the Above**

Q2. what is the difference between Autoencoders and RBMs?

- Autoencoders have less layeres than RBMS.
**Autoencoders use a deterministic approach, but RBMs use a stochastic approach.**- Autoencoders are used for supervised learning, but RBMs are used for unsupervised learning.
- All of the above

Q3. Which of the following problems cannot be solved by Autoencoders?

- Emotion Detection
**Time series prediction**- Image Reconstruction
- Dimensionality Reduction
- All of the above

Q4. What is a Deep Autoencoder?

- An Autoencoder stacked with over 1000 layers
- An Autoencoder with multiple input and output layers
- An Autoencoder stacked with Multiple Visible Layers
**An Autoencoder with Multiple Hidden Layers**- None of the Above

Q5. What is TRUE about Autoencoders?

- Used for Unsupervised Learning
- Used to Learn the Most important Features in Data
- Help to Reduce the Curse of Dimensionality
**All of the Above**

.

**Review: **

Based on our knowledge, we urge you to enroll in this course so you can pick up new skills from specialists. It will be worthwhile, we trust.