Machine Learning

10-701/15-781, Spring 2011

Carnegie Mellon University

Tom Mitchell


Date Lecture Topics Readings and useful links Handouts
Jan 11 Intro to ML
Decision Trees
Slides

video
  • Machine learning examples
  • Well defined machine learning problem
  • Decision tree learning
Mitchell: Ch 3
Bishop: Ch 14.4
The Discipline of Machine Learning
 
Jan 13 Decision Tree learning

Review of Probability

Annotated slides

video
  • The big picture
  • Overfitting
  • Random variables, probabilities
Andrew Moore's Basic Probability Tutorial
Bishop: Ch. 1 thru 1.2.3
Bishop: Ch 2 thru 2.2
HW1 out Jan 14
Jan 18
Probability and Estimation

Annotated slides

video
  • Bayes rule
  • MLE
  • MAP
Andrew Moore's Basic Probability Tutorial
Bishop: Ch. 1 thru 1.2.3
Bishop: Ch 2 thru 2.2
Jan 20
Naive Bayes

Annotated slides

video
  • Conditional independence
  • Multinomial Naive Bayes
Mitchell: Naive Bayes and Logistic Regression
Jan 25
Gaussian Naive Bayes
Slides
Annotated Slides

video
  • Gaussian Bayes classifiers
  • Document classification
  • Brain image classification
  • Form of decision surfaces
Mitchell: Naive Bayes and Logistic Regression
HW1 due
HW2 out
Jan 27
Logistic Regression

Slides
Annotated slides

video
  • Naive Bayes - the big picture
  • Logistic Regression: Maximizing conditional likelihood
  • Gradient ascent as a general learning/optimization method
Mitchell: Naive Bayes and Logistic Regression

Ng & Jordan: On Discriminative and Generative Classifiers, NIPS, 2001.
Feb 1
Linear Regression

Slides
Annotated slides

video
  • Generative/Discriminative models
  • minimizing squared error and maximizing data likelihood
  • bias-variance decomposition
  • regularization
Feb 3
Practical Issues

  • Feature selection
  • Overfitting
  • Bias-Variance tradeoff
Feb 8
Graphical models 1
Annotated slides

video
  • Bayes nets
  • representing joint distributions with conditional independence assumptions
HW3 out
Feb 15
Graphical models 2
slides
video
  • D-separation and Conditional Independence
  • Inference
  • Learning from fully observed data
  • Learning from partially observed data
Feb 17
Graphical models 3

annotated slides
video
  • EM
EM and HMM tutorial J. Bilmes
Feb 22
Graphical models 4
annotated slides
video
  • Mixture of Gaussians clustering
  • Learning Bayes Net structure - Chow Liu
Intro. to Graphical Models, K. Murphy
Graphical Models tutorial, M. Jordan
HW3 due
HW4 out
Feb 24
Computational
Learning
Theory
annotated slides
video
  • PAC Learning
Mitchell: Ch. 7
Mar 1
HW4 due
Mar 3 Midterm Exam
  • in class
  • open notes, open book, no internet
Midterm
Solution
Mar 15
Computational
Learning
Theory
annotated slides
video
  • Mistake bounds
  • Weighted Majority Algorithm
Mitchell: Ch. 7
Mar 17
Semi-Supervised
Learning
slides: CoTraining
NELL
video
  • CoTraining / Multi-view Learning
  • Never ending learning (NELL)
Mar 22
Hidden Markov Models
annotated slides

  • Markov models
  • HMM's and Bayes Nets
  • Other probabilistic time series models
Bishop Ch. 13
Mar 24
Neural Networks
slides

video
  • Non-linear regression
  • Backpropagation and Gradient descent
  • Learning hidden layer representations
Mitchell Ch. 4
Bishop Ch. 5
Project proposals due
Mar 29
Learning Representations I
slides

video
  • Artificial neural networks
  • PCA
Bishop Ch. 12 through 12.1
A Tutorial on PCA, J. Schlens
SVD and PCA, Wall et al.
Mar 31
Learning Representations II
slides

video
  • Deep belief networks
  • ICA
  • CCA
Deep Belief Nets paper, Hinton & Salakhutdinov
CCA Tutorial, M. Borga
Apr 5
Learning Representations III
slides

video
  • Fisher Linear Discriminant
  • Latent Dirichlet Allocation
  • Intro to Kernel Functions
Bishop Ch. 6.1 (required)
Bishop Ch. 6.2, 6.3 (optional)
Apr 7
Kernel Methods and SVM's
slides

video
  • Regression: Primal and Dual forms
  • Kernels and Kernel Regression
  • SVMs
Bishop Ch. 6.1
Bishop Ch. 7, through 7.1.2
Apr 12
SVM's II
slides

video
  • Maximizing the margin
  • Noise and soft margin SVM's
  • PAC learning and SVM's
  • Hinge loss, log loss, 0-1 loss
Bishop Ch. 7, through 7.1.2 Project midway report due
Apr 14 No CMU classes today
Apr 19
Active Learning
slides

video
Guest lecture: Dr. Burr Settles
  • Uncertainty sampling
  • Query by committee
Settles: Active learning survey
Apr 21
ML in Computational Biology
slides

video
Guest lecture: Prof. Ziv Bar-Joseph
Apr 26
Reinforcement Learning I
slides

video
  • Markov Decision Processes
  • Value Iteration
  • Q learning
Kaelbling et al.: Reinforcement Learning: A Survey
Apr 28
Reinforcement Learning 2
RL slides
Final study guide

video
  • Q learning in non-deterministic domains
  • RL as model for learning in animals
  • Final exam review
May 6 (Friday) Final Exam
  • 1-4pm
  • Location: Gates Hillman 4401
  • open notes, open book, no internet
Final study guide