Week 1 Flashcards
Supervised learning
learning has input and output. its a matter of finding a mapping from f(x) => y
Unsupervised learning
learning has no mapping. data is given as input and algorithm tries to find meaning in it
Problems where result is one of different labels
classification
Problems where result is a real value (dollars, weight)
regression
fit a straight line through data to make a prediction
linear regression
2 types of supervised learning problems
classification, regression
m
training examples
x’s
input variables / features
y’s
output variables / target variable
(x, y)
one training example (x,y data point)
(x^i, y^i)
i’th training example (x,y data point)
h : x –> y
hypothesis function (classifier ??): maps x's to y's
nother name for “input” variables
input features
nother name for “output” variable
target variable
A pair (x^i, y^i) is called a ___
training example
A set of training examples is called a ___
training set
X (uppercase) represents
the space of input values
Y (uppercase) represents
the space of output values
theta values are also called ___
parameters
square error cost function
1/2m * sum(h(x)^i - y^i)^2
J(theta0, theta1…)
cost function we want to minimize
