Time
|
Content
|
Lecture Notes
|
Aug 25
|
Lecture 1: Course Description
and Policy. Introduction to learning from data
|
PDF
slides |
Aug 27
|
Lecture 2: Curve Fitting, Error
Function, Overfitting, Regularization, Bayesian Predictive
Distribution, Brief review of probability theory, decision
theory, and concepts in information theory (entropy, mutual information
and KL divergence) |
PDF
slides
|
Sept 1
|
Lecture 3: Brief review of
probability theory, decision
theory, and concepts in information theory (entropy, mutual information
and KL divergence) |
PDF
slide
|
Sept 3
|
Lecture 4: Maximum
Likelihood Estimation for Bernoulli and Mulitnomial Distributions, Beta
and Dirichilet Distributions, Conjugate Priors, Bayesian Estimation and
Predictive Posterior Distributions |
PDF
slides
|
Sept 8
|
Lecture 5: t-distributions,
mixture of Gassuians, Exponential Family,
Natural Parameters, Convexity of Normalization Coefficeint,
Conjugate Prior for Exponential Family |
PDF
slides
|
Sept 10
|
Lecture 6: Noinformative Prior,
Nonparametric Methods, Parzen Window, K-Nearest-Neigbhors Classification |
PDF
slides
|
Sept 15
|
Lecture 7: Linear Regression,
Ridge Regression, Lasso, Visualization of Regularized Regression,
Bayesian Regression, Model Comparison, Bayes Factor |
PDF
slides |
Sept 17
|
Lecture 8: Evidence
Framework, Empriical Bayes, Linear Classification Models, Margin,
Fishe'sr Linear Discriminant |
PDF
slides |
Sept 22
|
Lecture 9: Fishe'sr Linear
Discriminant, Perceptron,
Generalized Linear Models, Generative Models for classification, Linear
Gaussian Classifier |
PDF
slides |
Sept 24
|
Lecture 10: Logistiic
Regression, Probit Regression, Newton-Raphson Optitimization |
PDF
slides |
Sept 29
|
Lecture 11: Laplace
Approxmation, BIC model selection, Bayesian Logistic
Regression, Kernel Methods, Gram matrix
|
PDF
slides |
Oct 1
|
Lecture 12: Dual
Represenation, Kernel Ridge Regression, Combining
Generative and Discriminative Models by Kernels, Fisher Kernel,
Kernel PCA |
PDF
slides |
Oct 6
|
Lecture 13: Gaussian Processes |
PDF
slides |
Oct 8
|
Lecture 14: Gaussian
Process Regression, Automatic Relevance Determination |
PDF
slides |
Oct 16
|
Lecture 15: Gaussian
Process Classification |
PDF
slides |
Oct 20
|
Lecture 16: Midterm review
|
PDF
slides |
Oct 22
|
Lecture 17: Midterm
|
|
Oct 27
|
Lecture 18: Support Vector
Machines for Classification, Introduction to Lagrange Multipliers
|
PDF
slides |
Oct 29
|
Lecture 19: Support Vector
Machines for Classification and Regression, SVM and
Regularization, Graphical models, Bayesian networks
|
PDF
slides |
Nov 2
|
Lecture 20: Conditional
Independence, Explaining
Away Effect, D-separation,
Markvo Chains, Markov Random Fields, ICM, Moralization |
PDF
slides |
Nov 4
|
Lecture 21: Inference on chains,
Factor graphs, Sum-product algorithm,
|
PDF
slides |
Nov 10
|
Lecture 22: Sum-product
algorithm, Loopy belief propagation,
Junction Tree Algorithm, Maximum marginals vs
maximum
joint distribution, Max-sum algorithm |
PDF
slides |
Nov 12
|
Lecture 23: K-means
clustering, Vector
quantization,, K-medoids clustering, Mixture of Gaussians, Expectation
maximization (EM), Lower Bounds in EM |
PDF
slides |
Nov 17
|
Lecture 24: Hidden Markvo
Models, forward-backward algorithm, EM for learning HMM parameters,
Viterbi Algorithm |
PDF
slides |
Nov 19
|
Lecture 25: Linear state space
models, Kalman filtering and
smoothing, Importance Sampling |
PDF
slides |
Nov 24
|
Lecture 26: Rejection Sampling,
Importance Sampling, Detailed Balance, Metroplis-hasting algorithm,
Gibbs sampling |
PDF
slides |