Advanced Machine Learning

Question 1

What is Deep Learning?

Answer 1

Deep learning is a subset of machine learning that has become increasingly popular in recent years due to its performance along with improved computer speed.

Question 2

What is the process of neural networks?

Answer 2

1. Take our inputs and multiply each input by a weight.
2. Add the results from step 1 together.
3. Add a value (called bias) to the result from step 2
4. Pass result from step 3 through an activation function.
5. The result from step 4 is then passed to the next layer, where the processed is repeated.

Question 3

What is Node?

Answer 3

One unit that process incoming information and produces a single output number

Question 4

What is layer?

Answer 4

A collection of nodes working in parallel

Question 5

What is Input layer?

Answer 5

The first set of nodes that work on the features of the input sample

Question 6

What is Output layer?

Answer 6

The last set of nodes that output a prediction

Question 7

What is Forward Propagation?

Answer 7

The processing and passing of information forward through all layers to produce an output prediction

Question 8

What is cost function?

Answer 8

The difference between the predictions of all samples and their true labels

Question 9

What is Backward Propagation?

Answer 9

The process of updating the weights of each node to reduce the cost function

Question 10

What is Epoch?

Answer 10

The process of completing one forward Propagation step on each sample in the training set and updating the weights of each node with backward Propagation is called an epoch

Question 11

What is activation functions?

Answer 11

They can create non-linear functions to predict class or values

Question 12

What is sigmoid?

Answer 12

The sigmoid function maps all output values of the node to a value between 0 and 1

Question 13

What is Tanh (or hyperbolic tangent)?

Answer 13

Tanh maps the output of a node to a value between -1 and 1, as shown above on the y axis of the second plot

Question 14

Wha is ReLu (Rectified Linear Unit)?

Answer 14

It maps all negative outputs from a node to 0 and returns all positive outputs as is (linear function)

Question 15

What is Output layer Activation functions?

Answer 15

The exception is the output layer. Specific activation functions are required for a model to produce specific kinds of outputs

Question 16

What is Linear (regression models)?

Answer 16

If the model should have a continuous output that can produce positive or negative. A Linear activation function simply returns the value that was passed to it. g(z) =z.

Question 17

What is sigmoid (binary classification)?

Answer 17

If the model should only output 0 or a 1, such as in binary classification problems

Question 18

What is softtmax (multiclass classification models)?

Answer 18

If the model should return the most likely candidate from a finite list of options, such as a multiclass classification problem

Question 19

What is ythat?

Answer 19

The prediction of a model for all samples in X.

Question 20

What is cost?

Answer 20

The combination of all of a model’s errors on all samples

Question 21

What is cost function or loss function?

Answer 21

The difference between the predictions of all samples and their true labels

Question 22

What is Gradient Descent?

Answer 22

Changing weights to reduce the cost function

Question 23

What is Keras?

Answer 23

Keras is an API that acts as an interface for tensorflow, a popular choice for building networks, especially as you are first learning

Question 24

What is Bias?

Answer 24

Is is related to underfitting

Question 25

What is Variance?

Answer 25

Is it related to overfitting

Question 26

What is CNN?

Answer 26

Convolutional Nerual Networks (CNNs) revolutionized computer vision because of how well they do at working with image data.