Syllabus and (TENTATIVE) Lecture Schedule

 

Date	Lecture	Topics	Readings and useful links	Handouts
Module 1: Introduction to Machine Learning
Mon Jan 12	Overview	Overview of Machine Learning Decision tree learning algorithm lecture slides for Jan 12 and Jan 14	Mitchell: Chap 3 (Decision Trees) The Discipline of Machine Learning, T. Mitchell, 2006. Decision tree applet (courtesty of Univ. of Alberta, AAAI)	HW1 out
Wed Jan 14	Decision trees	Decision trees overfitting pruning cross validation Entropy lecture slides for Jan 12 and Jan 14	Mitchell: Chap 3 Bishop: Chap 1.6 come to Jan 15 recitation on Decision trees and entropy, 5-6pm, NSH 1305
Mon Jan 19	NO CLASS	Martin Luther King Day
Wed Jan 21	Probability	Probability Axioms, distributions, Bayes Rule lecture slides	Bishop 1.2	HW1 due HW2 out
Mon Jan 26		Maximum likelihood and MAP estimators, Conditional independence lecture slides
Module 2: Supervised Learning
Wed Jan 28	Bayes Classifiers	Naive Bayes classifier conditional independence lecture slides	Mitchell chapter on Naive Bayes and Logistic Regression
Mon Feb 2		Gaussian Naive Bayes text classification image classification lecture slides		HW2 due
Wed Feb 4	Logistic regression	Logistic Regression maximizing conditional likelihood gradient descent generative and discriminative  lecture slides	Required reading: Mitchell chapter on Naive Bayes and Logistic Regression Optional reading: On Discriminative and Generative Classifiers, Ng and Jordan, NIPS, 2001.	HW3 out
Mon Feb 9		Practical issues estimating accuracy confidence intervals cross validation lecture slides
Wed Feb 11	Regression	Linear regression feature selection regularization Regression and its probabilistic interpretation lecture slides
Mon Feb 16	Bayesian networks	Bias-Variance decomposition of error Bayes nets Directed networks representing joint distribution conditional indep assumptions lecture slides
Wed Feb 18		Inference and Supervised learning of Bayes Net parameters lecture slides		HW3 due HW4 out
Mon Feb 23		D-Separation and conditional independence EM and Learning from partly-observed data lecture slides	Required reading on D-separation: Bishop section 8.2
Wed Feb 25		Mixture of Gaussians clustering Learning parameters and network structure lecture slides	EM Mixture of Gaussians applet
Mon Mar 2	Midterm review	slides from review session		HW4 due
Wed Mar 4		Midterm Exam (solutions)	open book, open notes, no computers
Mar 9 Mar 11		SPRING BREAK
Mon Mar 16	Learning theory I	Probably approximately correct learning lecture slides	Recommended reading: Mitchell, Ch. 7
Wed Mar 18	Learning theory II	VC dimension, mistake bounds lecture slides	Recommended reading: Mitchell, Ch. 7
Mon Mar 23	Support Vector Machines	SVM's Margin-based methods Kernel trick lecture slides	guest lecture by Professor Ziv Bar-Joseph	project proposals due at start of class
Wed Mar 25		SVM's part II	guest lecture by Professor Ziv Bar-Joseph
Mon Mar 30	Neural networks	Artificial neural networks lecture slides	recommended reading: Mitchell Ch. 4	HW5 out
Wed Apr 1	Semi-supervised learning I	EM-based Changing the objective Metric regularization lecture slides	recommended readings: Text Classification from Labeled and Unlabeled Documents using EM Metric-Based Methods for Adaptive Model Selection and Regularization
Mon Apr 6	Semi-supervised learning II	Co-training and coupling functions lecture slides		HW5 due
Wed Apr 8	Dimensionality reduction I	Discovering lower dimensional representations PCA, SVD lecture slides	recommended readings: PCA tutorial by Schlens PCA tutorial by Wall
Mon Apr 13	Student presentations	2 minute summary of your project midway RESULTS		project midway reports due
Wed Apr 15	Dimensionality reduction II	CCA, latent variables, and topic models. Applications to fMRI, text, social network analysis see Apr 8 slides.
Mon Apr 20	class cancelled
Wed Apr 22	Time series data	Comparison of PCA and Neural Nets for face image analysis slides (courtesy of Portia Taylor) Hidden Markov Models  lecture slides	recommended HMM readings: Bishop Ch. 13 Rabiner paper on HMMs
Mon Apr 27	ML Applications	ML for Computational Biology	guest lecture by Prof. Ziv Bar-Joseph
Wed Apr 29	Poster session	NOTE SPECIAL TIME: 3-5pm SPECIAL LOCATION: NSH Atrium	No lecture today.	poster presentation
Mon May 4	no class		Submit paper copy of your report to Sharon Cavlovich by 5pm. ALSO submit email copy to the instructor for your project.	final project reports due.
Fri May 8 8:30am-11:30am room: HH B103		Final Exam	open book, open notes, no computers, no network

 

© 2009 Tom Mitchell @ School of Computer Science, Carnegie Mellon University