Professional Documents
Culture Documents
1. Foundations
4. Clustering 5. Recommender
2. Regression 3. Classification
& Retrieval Systems
6. Capstone
ML
Data Regression Intelligence
Method
Input x:
features derived Learn xày
from data
relationship Predict y:
continuous “output” or
“response” to input
5
©2015
Emily
Fox
&
Carlos
Guestrin
Machine
Learning
Specializa0on
Salary after ML specialization
hard work
Output y
very sad very happy
Inputs x are
brain region
intensities
ML
Data Regression Intelligence
Method
$ = ??
price ($)
$ $
$ (y)
+ house house size
attributes (x)
house size x
Define goodness-of-fit
metric for each possible line
house size
Gradient descent algorithm
better fit
Get estimated
parameters
slope
- interpret
- use to form
intercept predictions
14
©2015
Emily
Fox
&
Carlos
Guestrin
Machine
Learning
Specializa0on
Module 2: Multiple Regression
y
Fit more complex
price ($)
relationships than
just a line
- Square feet
x[2] - # bathrooms
- # bedrooms
- Lot size
- Year built
house size x[1]
- …
15
©2015
Emily
Fox
&
Carlos
Guestrin
Machine
Learning
Specializa0on
Module 3: Assessing Performance
y y
price ($)
price ($)
R F I T
house size x VE size
Ohouse x
Measures of error:
- Training
- Test
- True (generalization)
price ($)
R F I T
house size x VE size
Ohouse x
Bias-variance
tradeoff
price ($)
R F I T
house size x VE size
Ohouse x
Cross validation
Valid
set
error2(λ) ŵλ(2)
19
©2015
Emily
Fox
&
Carlos
Guestrin
Machine
Learning
Specializa0on
Module 5: Feature Selection
& Lasso Regression
Lot
size
Dishwasher
Single
Family
Garbage
disposal
Year
built
Microwave
$?
Last
sold
price
Range
/
Oven
Last
sale
price/sqM
Refrigerator
Finished
sqM
Washer
Unfinished
sqM
Dryer
Finished
basement
sqM
Laundry
loca0on
#
floors
Hea0ng
type
Flooring
types
Je]ed
Tub
Parking
type
Deck
Useful for efficiency Parking
amount
Fenced
Yard
Cooling
Lawn
of predictions and Hea0ng
Garden
interpretability Exterior
materials
Roof
type
Sprinkler
System
…
Structure
style
20
©2015
Emily
Fox
&
Carlos
Guestrin
Machine
Learning
Specializa0on
Module 5: Feature Selection
& Lasso Regression
1.5
1 f(x0)
0.5
−0.5
$ = ???
−1
• Gradient descent
Algorithms • Coordinate descent