Lectures / Schedule

Tentative Schedule (in progress)

Supervised Learning
R Jan 11
Introduction and course overview;
Binary classification and Bayes error
[slides] [lecture notes]
Bishop Ch 1
Duda, Hart & Stork, Pattern Classification
T Jan 16 [Class postponed]      
R Jan 18 [Class postponed]      
T Jan 23
Generative probabilistic models
for classification
Bishop Ch 4
  PS1 out
R Jan 25
Discriminative probabilistic models
for classification
Bishop Ch 4
F Jan 26 Least squares regression Bishop Ch 3    
On your
Review: Constrained optimization
and Lagrangian duality
  Boyd & Vandenberghe, Convex Optimization  
T Jan 30 Support vector machines Bishop 7.1    
R Feb 1 Kernel methods Bishop Ch 6    
F Feb 2 Neural networks/deep learning Bishop Ch 5    
T Feb 6
Decision trees;
Nearest neighbor methods
Bishop 14.4, 2.5.2
PS1 due;
PS2 out
R Feb 8 No class      
T Feb 13 Boosting      
R Feb 15 No class      
T Feb 20
Model selection;
Generalization error bounds/VC dimension;
Structural risk minimization
    PS2 due;
PS3 out
Unsupervised Learning
R Feb 22 Principal component analysis      
T Feb 27 Midterm exam (in class)      
R Mar 1 Auto-encoding neural networks      
T Mar 6 No class (Spring Break)      
R Mar 8 No class (Spring Break)      
T Mar 13
k-means clustering;
Mixture models and EM algorithm
PS3 due;
PS4 out;
proposals due
Probabilistic Graphical Models
R Mar 15 Hidden Markov models      
T Mar 20 Hidden Markov models      
R Mar 22 Bayesian networks      
T Mar 27
Markov networks; variable elimination
PS4 due;
PS5 out
R Mar 29
Message-passing algorithms:
sum-product, max-sum
Other Learning Frameworks
T Apr 3 Online learning      
R Apr 5
Semi-supervised learning;
Active learning
T Apr 10
Reinforcement learning
PS5 due;
PS6 out
Additional/Advanced Topics
R Apr 12 Structured prediction      
T Apr 17 Complex performance measures      
R Apr 19 Statistical consistency and PAC learning      
Home Stretch
T Apr 24
Project poster presentations
PS6 due;
reports due
TBA Final exam