Machine Learning (Coursera, Andrew Ng)

Question 1

Usages for Machine Learning?

Answer 1

Applications cant program by hand

Question 2

Examples of Database mining?

Answer 2

Making sense of web-click data, medical records

Question 3

Making sense of web-click data is about?

Answer 3

Data Mining

Question 4

Making sense of medical records is about?

Answer 4

Data Mining

Question 5

Amazon/Netflix recommendations is an example of?

Answer 5

Self-customizing programs

Question 6

Field of study that gives computers the ability to learn without being explicitly programmed?

Answer 6

Machine Learning

Question 7

A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, measured by P, improves with experience E?

Answer 7

Well-posed Learning Problem

Question 8

What is Machine Learning?

Answer 8

Field of study that gives computers the ability to learn without being explicitly programmed

Question 9

What is Well-posed Learning Problem?

Answer 9

A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, measured by P, improves with experience E

Question 10

Two broad classifications in ML?

Answer 10

Supervised Learning - Unsupervised Learning

Question 11

What problems Supervised Learning solves?

Answer 11

Given the data to learn on - can we predict the future results?

Question 12

What problems Unsupervised Learning solves?

Answer 12

Given a dataset - can we find some structure?

Question 13

The ML Problems?

Answer 13

Regression, Classification, Clustering

Question 14

Regression?

Answer 14

Predict continuous value output

Question 15

Classification?

Answer 15

Predict discrete value output

Question 16

Clustering?

Answer 16

Grouping the set of inputs together

Question 17

To what ML Problems corresponds - Price Prediction?

Answer 17

Regression

Question 18

In Regression problem what does m stand for?

Answer 18

Number of training examples

Question 19

In Regression problem what does x stand for?

Answer 19

Input feature

Question 20

In Regression problem what does y stand for?

Answer 20

Output variable

Question 21

In Regression problem what does h stand for?

Answer 21

Hypothesis function

Question 22

Examples for Regression Application?

Answer 22

Price prediction

Question 23

Formula for linear regression?

Answer 23

h(x) = theta0 + theta1 * x

Question 24

Purpose of cost function?

Answer 24

Estimate correctness of our hypothesis function

Question 25

Squared error cost function?

Answer 25

https://www.codecogs.com/eqnedit.php?latex=J(%5Ctheta_0%2C%20%5Ctheta_1)%20%3D%20%5Cfrac%7B1%7D%7B2m%7D%20%5Csum_%7Bi%3D1%7D%5Em(%20%5Chat%7By%7D_i%20-%20y_i%20)%5E2%20%3D%20%5Cfrac%7B1%7D%7B2m%7D%20%5Csum_%7Bi%3D1%7D%5Em(%20h_%5Ctheta%20(x_i)%20-%20y_i)%5E2

Question 26

How we choose parameters for linear regression?

Answer 26

In order to minimize the difference between hypothesis and real value

Question 27

What is Gradient Descent?

Answer 27

Minimization Algorithm

Question 28

What does “Batch Gradient Descent” mean?

Answer 28

All training examples are analyzed on each step

Question 29

How does Gradient Descent Algorithm look like?

Answer 29

repeat until convergence {

} where j=0,1 represents the feature index number. (Simultaneous assignment)

Question 30

What happens if learning rate of Gradient Descent is too small?

Answer 30

It might take a lot of steps before convergence

Question 31

What happens if learning rate of Gradient Descent is too big?

Answer 31

It may fail to converge, or even diverge

Question 32

Can Gradient Descent converge with fixed learning rate?

Answer 32

Yes

Question 33

Why Gradient Descent converges with fixed learning rate?

Answer 33

Because derivative is changing as we approach the minimum