# Deep Learning and Reinforcement Learning Week 2

Course Name: Deep Learning and Reinforcement Learning

Course Link: Deep Learning and Reinforcement Learning

### Practice: Introduction to Neural Networks

Q1. Select the method or methods that best help you find the same results as using matrix linear algebra to solve the equation θ=(XTX)-1XTy
Use scikit-learn to build a linear regression model
Train a neural network model
All the above

Q2. (True/False) Neurons can be used as logic gates
True
False

Q3. (True/False) The feed-forward computation of a neural network can be thought of as matrix calculations and activation functions.
True
False

These are answers of Deep Learning and Reinforcement Learning Week 2 Quiz

### Practice: Keras Library

Q1. Building a Neural Network with the Sequential API in Keras implies that each layer
can connect only to subsequent layers.
after the first layer must be fully-connected.
can connect to only the previous and next layers.
can be connected to at most two subsequent layers.

Answer: can connect to only the previous and next layers.

Q2. An epoch in estimating a Deep Learning model refers to
subsamples that make up the training data.
iterations required to achieve convergence.
computational complexity of the estimating procedure.
the number of times the entire input data set is used by the model.

Answer: the number of times the entire input data set is used by the model.

These are answers of Deep Learning and Reinforcement Learning Week 2 Quiz

Q3. An advantage of the Sigmoid activation function over the step activation function is:
the ability to generate nonlinear outcomes.
sharper changes as the loss becomes positive.
improved backpropagation due to nonzero gradients.
significantly different gradients for very high values.

These are answers of Deep Learning and Reinforcement Learning Week 2 Quiz

### Final Quiz

Q1. The backpropagation algorithm updates which of the following?
The losses only.
The activations only.
The parameters and activations.
The parameters only.

Q2. What of the following about the activation functions is true?
They add non-linearity into the model, allowing the model to learn complex pattern.
They tell us about how computationally expensive a neural network is.
They evaluate how well the model has performed on the training data.
They are optimization algorithms that update values of the model parameters.

Answer: They add non-linearity into the model, allowing the model to learn complex pattern.

These are answers of Deep Learning and Reinforcement Learning Week 2 Quiz

Q3. What is true regarding the backpropagation rule?
The actual output is determined by computing the output of neurons in each hidden layer
It prevents overfitting
It can be used to update the hyperparameters of a neural network
It is a feed forward neural network.

Answer: The actual output is determined by computing the output of neurons in each hidden layer

Q4. Which option correctly lists the steps to build a linear regression model using Keras?
1. Use `fit()` and specify the number of epochs to train the model for.
2. Create a Sequential model with the relevant layers.
3. Normalize the features with ` layers.Normalization()` and apply `adapt()`.
4. Compile using `model.compile()` with specified optimizer and loss.

3, 2, 4, 1
3, 2, 1, 4
3, 1, 2, 4
2, 4, 3, 1

These are answers of Deep Learning and Reinforcement Learning Week 2 Quiz

Q5. (True/False) Keras provides one approach to build a model: by defining a Sequential model.
True
False

These are answers of Deep Learning and Reinforcement Learning Week 2 Quiz

More Coursera Courses: http://progiez.com/coursera

 The content uploaded on this website is for reference purposes only. Please do it yourself first.