ece#5424:#introduction#to#...
TRANSCRIPT
ECE 5424: Introduction to Machine Learning
Stefan LeeVirginia Tech
Topics: – Regression
Readings: Barber 17.1, 17.2
Administrativia• HW1
– 39 Submissions– 58 Students Enrolled– 19 MIA (?????)
• Project Proposal– Due: 09/21, 11:55 pm (LESS THAN A WEEK!)– <= 2pages, NIPS format
• HW2 – Out today– Due on Wednesday 09/28, 11:55pm– Please please please please please start early– Implement Linear Regression, Naïve Bayes, Logistic Regression
(C) Dhruv Batra 2
Recap of last time
(C) Dhruv Batra 3
Learning a Gaussian• Collect a bunch of data
– Hopefully, i.i.d. samples– e.g., exam scores
• Learn parameters– Mean– Variance
(C) Dhruv Batra 4
MLE for Gaussian• Prob. of i.i.d. samples D={x1,…,xN}:
• Log-likelihood of data:
Slide Credit: Carlos Guestrin(C) Dhruv Batra 5
Your second learning algorithm:MLE for mean of a Gaussian
• What’s MLE for mean?
Slide Credit: Carlos Guestrin(C) Dhruv Batra 6
Learning Gaussian parameters
• MLE:
(C) Dhruv Batra 7
• Conjugate priors– Mean: Gaussian prior– Variance: Inverse Gamma or Wishart Distribution
• Prior for mean:
Slide Credit: Carlos Guestrin(C) Dhruv Batra 8
Bayesian learning of Gaussian parameters
MAP for mean of Gaussian
Slide Credit: Carlos Guestrin(C) Dhruv Batra 9
New Topic: Regression
(C) Dhruv Batra 10
1-NN for Regression• Often bumpy (overfits)
(C) Dhruv Batra 11Figure Credit: Andrew Moore
(C) Dhruv Batra 12Slide Credit: Greg Shakhnarovich
(C) Dhruv Batra 13Slide Credit: Greg Shakhnarovich
(C) Dhruv Batra 14Slide Credit: Greg Shakhnarovich
Linear Regression• Demo
– http://hspm.sph.sc.edu/courses/J716/demos/LeastSquares/LeastSquaresDemo.html
(C) Dhruv Batra 15
(C) Dhruv Batra 16Slide Credit: Greg Shakhnarovich
Plan for Today• Regression
– Linear Regression Recap• Some matrix calculus review
– Connections with Gaussians• The outlier problem L
– Robust Least Squares
(C) Dhruv Batra 17
(C) Dhruv Batra 18Slide Credit: Greg Shakhnarovich
(C) Dhruv Batra 19Slide Credit: Greg Shakhnarovich
(C) Dhruv Batra 20Slide Credit: Greg Shakhnarovich
(C) Dhruv Batra 21Slide Credit: Greg Shakhnarovich
(C) Dhruv Batra 22Slide Credit: Greg Shakhnarovich
(C) Dhruv Batra 23Slide Credit: Greg Shakhnarovich
• Why sum squared error???• Gaussians, Watson, Gaussians…
But, why?
24(C) Dhruv Batra
(C) Dhruv Batra 25Slide Credit: Greg Shakhnarovich
MLE Under Gaussian Model• On board
(C) Dhruv Batra 26
Is OLS Robust?• Demo
– http://www.calpoly.edu/~srein/StatDemo/All.html
• Bad things happen when the data does not comefrom your model!
• How do we fix this?
(C) Dhruv Batra 27
Robust Linear Regression• y ~ Lap(w’x, b)
• On board
(C) Dhruv Batra 28
0 0.2 0.4 0.6 0.8 1−6
−5
−4
−3
−2
−1
0
1
2
3
4Linear data with noise and outliers
least squareslaplace
−3 −2 −1 0 1 2 3−0.5
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
L2
L1
huber