Time
|
|
Lecture Notes
|
Jan 11
|
Lecture 1: Course description
and policy.
|
PDF
slides |
Jan 13
|
Lecture 2: Curve fitting, error
function, overfitting, regularization, and maximum likelihood
stimation for Gaussian distributions
|
PDF
slides
|
Jan 18
|
Lecture 3: Bayesian predictive
distributions, brief review of probability theory, decision
theory, and elements in information theory (entropy, mutual information
and KL divergence) |
PDF
slides |
Jan 20
|
Lecture 4: Entropy, mutual
information, KL divergence, maximum likelihood and Bayesian estimation
for Bernoulli distributions, Beta distributions, conjugate priors
|
PDF
slides |
Jan 25
|
Lecture 5: Dirichilet
distributions, ML and Bayesian estimation of multinomial distributions
|
PDF
slides |
Jan 27
|
Lecture 6: ML and Bayesian
estimation of Gaussian distributions
|
PDF
slides |
Feb 1
|
Class canceled due to the winter
storm
|
|
Feb 3
|
Lecture 7: t-distributions,
mixture of Gaussians, exponential family, sufficient statistics, ML and
Bayesian estimation for exponential family
|
PDF
slides |
Feb 8
|
Lecture 8: Non-informative
prior, Nonparametric methods, Parzen windows, K-nearest neighbors
|
PDF
slides |
Feb 10
|
Lecture 9: K-nearest neighbors,
Linear Regression with basis functions |
PDF
slides |
| Feb 15 |
Lecture 10: Ridge regression,
Lasso
|
PDF
slides |
Feb 22
|
Lecture 11: Bayesian regression,
Model comparison, Bayes factor, empirical Bayes
|
PDF
slides |
Feb 24
|
Lecture 12: Empriical Bayes,
Linear Classification Models, Margin,
Fishe'sr Linear Discriminant |
PDF
slides |
March 1
|
Lecture 13: Fishe'sr Linear
Discriminant, Perceptron, Generative Models for classification, Linear
Gaussian Classifier, Logistiic
Regression |
PDF
slides |
March 3
|
Lecture 14: Probit Regression,
Newton-Raphson Optitimization, Laplace's approximation, Bayesian
logistic regression, Kernel methods |
PDF
slides |
March 8
|
Lecture 15: Kernel construction,
Kernel ridge regression, Kernel PCA
|
PDF
slides |
March 8
|
Lecture 16: Introduction to
Gaussian processes, consistency condition
|
PDF
slides |
March 15,17
|
Spring Break
|
|
March 22
|
Lecture 17: Gaussian process
regression, automatic relevance determination, Gaussian process
classfication
|
PDF
slides |
March 24
|
Lecture 18: Midterm review
|
PDF
slides |
March 29
|
Lecture 19: Support Vector
Machines for cassification, Introduction to Lagrange multipliers |
PDF
slides
|
April 5
|
Lecture 20: Karush-Kuhn-Tucker
(KKT) condition, Support vector machines for classification and
regression
|
PDF
slides |
April 7
|
Lecture 21: SVM regression,
Bayesian networks, conditional independence, explaining away effect,
D-separation |
PDF
slides
|
April 14
|
Lecture 22: Markov
blankets, Markov random fields, inference on
chain
|
PDF
slides |
April 19
|
Lecture 23: Factor graphs, the
sum-product algorithm (belief propagation), the junction tree algorithm
|
PDF
slides |
April 21
|
Lecture 24: The max-sum
algorithm, K-mean clustering, mixture of Gaussians, expectation
maximization, lower-bound interpretation of EM
|
PDF
slides
|
April 26
|
Lecture 25: EM, Hidden Markov
models, and course review
|
PDF
slides
|