Week 1

Question 1

Supervised learning

Answer 1

learning has input and output. its a matter of finding a mapping from f(x) => y

Question 2

Unsupervised learning

Answer 2

learning has no mapping. data is given as input and algorithm tries to find meaning in it

Question 3

Problems where result is one of different labels

Answer 3

classification

Question 4

Problems where result is a real value (dollars, weight)

Answer 4

regression

Question 5

fit a straight line through data to make a prediction

Answer 5

linear regression

Question 6

2 types of supervised learning problems

Answer 6

classification, regression

Question 7

m

Answer 7

training examples

Question 8

x’s

Answer 8

input variables / features

Question 9

y’s

Answer 9

output variables / target variable

Question 10

(x, y)

Answer 10

one training example (x,y data point)

Question 11

(x^i, y^i)

Answer 11

i’th training example (x,y data point)

Question 12

h : x –> y

Answer 12

hypothesis function (classifier ??):
maps x's to y's

Question 13

nother name for “input” variables

Answer 13

input features

Question 14

nother name for “output” variable

Answer 14

target variable

Question 15

A pair (x^i, y^i) is called a ___

Answer 15

training example

Question 16

A set of training examples is called a ___

Answer 16

training set

Question 17

X (uppercase) represents

Answer 17

the space of input values

Question 18

Y (uppercase) represents

Answer 18

the space of output values

Question 19

theta values are also called ___

Answer 19

parameters

Question 20

square error cost function

Answer 20

1/2m * sum(h(x)^i - y^i)^2

Question 21

J(theta0, theta1…)

Answer 21

cost function we want to minimize