CMPE 58K, Bayesian Statistics and Machine Learning

Fall 2009

Instructor: Ali Taylan Cemgil
Bogaziçi University, Department of Computer Engineering, Istanbul, Turkey

Announcements

Note the schedule changes 
14 Jan 2010 cmpe58K Presentations

Session 1
9:30 AYCA KUNDAK                     
9:50 DERYA ERHAN 
10:10 FATMA BASAK AYDEMIR 
10:30 BORA CAGLAYAN 
Break 10 min

Session 2
11:00 ALPARSLAN	YILDIZ
11:20 COSKUN MERMER 
11:40 MEHMET YILMAZ 
12:00 OGUZ YILMAZ 

Lunch 70 min

Session 3
13:30 CAN KAVAKLIOGLU
13:50 FAHRI SERHAN DANIS
14:10 ERINC DIKICI
14:30 DOGAC	BASARAN

Break 10 min

Session 4
15:00 NEZIH ERGUN OZKUCUR
15:20 UMIT ISLAK
15:40 SERKAN CIMEN
16:00 UMUT FIRAT
Take Home Final Deadline 11.01.2010 (Passed)

Catalog Information (pdf)

Slides

Lecture00  Course Structure
Lecture01  Introduction, Probability tables,
Lecture02  Graphical Models, Independence, Conditional Independence,  Additional slides from Bishop 
Lecture03  Bayesian Learning, Applications
Lecture04  Exponential Family, Unvariate Probability Distributions, Conjugate Priors
Lecture05  Multivariate Distributions: Gaussian, Multivariate Bernoulli
Lecture06  Sequential Data, Markov models, Hidden Markov Models (HMM's), Inference, Forward Backward algorithm
Lecture07  Approximate inference, Variational Methods: Variational Bayes, ICM, Expectation Maximisation (EM), Variational Methods, AR Model Example
Lecture08  Correction Smoother, Junction tree algorithm
Lecture09  Linear Dynamical Systems (LDS's), Inference in LDS, Kalman Filter and Smoother
Supplement  Changepoint models

Further Topics

not covered in detail:
 Undirected graphical models, Markov Random fields, Boltzman machines
 Exact Inference, Belief Propagation, Sum Product algorithm
 Gaussian Processes
 Construction of Probabilistic models, Hierarchical Modeling, Latent variable models
 Switching state space models

not covered at all:
 Mean field, Gaussian Mixtures, PPCA
 Dirichlet, Wishart Distributions
 Bayesian learning in HMM's and LDS's,
 Advanced topics: Nonparametric Bayesian models, Dirichlet Process Mixtures,
 Modeling Relational Data, Probabilistic SVD, Bayesian Matrix factorisations, Latent Dirichlet allocation

Lecture Outline, Assignments

DateTopicReadingAssignmentSolutions
Oct 01, Thu Course Structure, Bayes Theorem, Applications,
Oct 08, Thu Graphical Models, DAG's, MRF's and Factor Graphs, Conditional Independence Chapter 1 and 8.1.1-8.1.3, 8.2.1 from Bishop
A short introduction to graphical models by Kevin Murphy 		HW1
Oct 15, Thu D-separation, Blocking, Conditional Independence, Applications, Sequential Data,
Oct 22, Thu Inference in Hidden Markov Models (HMM's), Forward Backward algorithm HW2
Oct 29, Thu Cumhuriyet Bayrami
Nov 05, Thu Bayesian Learning,
Nov 12, Thu Probability Distributions, Exponential Family, Conjugate Priors,
Nov 19, Thu Multivariate Gaussians, Construction of Probabilistic models, Hierarchical Modeling,
Nov 26, Thu Changepoint Models (CPM), Exact Inference in CPM, Switching state space models,
Dec 03, Thu Midterm Review, Approximate inference, Variational Methods, Changepoint Models Tutorial HW3
Dec 10, Thu Midterm I
Dec 17, Thu Mean field, Variational Bayes, ICM, Expectation Maximisation (EM),
Dec 24, Thu Linear Dynamical Systems (LDS's), Inference in LDS, Kalman Filter and Smoother, Bayesian learning in HMM's and LDS's
Dec 31, Thu Exact Inference, Junction Tree algorithm, Belief Propagation, Sum Product algorithm

bayes.jpg

Past Announcements