Syllabus for Machine Learning 10-601 in Fall 2014

This is the syllabus for Machine Learning 10-601 in Fall 2014.

Schedule

Schedule for 10-601 in Fall 2013

Date of lecture	Topic	Lecturer	Assignment
Tu 9/2	Overview and Intro to Probability	William
Th 9/4	Classification and K-NN	William	slides will be updated	-	Tu 9/9	Decision Trees, and Rule Learning	William	slides will be updated
Th 9/11	The Naive Bayes algorithm	William
Tu 9/16	Linear Regression	William	slides will be updated
Th 9/18	Logistic Regression	William
Mo 9/22		William also lectures in Ziv's class		Tu 9/23	The Perceptron algorithm	William
W 9/18	SVMs and Margin Classifiers	William	HW3: Logistic Regression Download: data (due Sept. 25th via Autolab) example solution
W 9/25	Neural networks and Deep Belief Networks	Eric	HW4: Linear Regression Download: data (due Oct. 2nd (Before lecture) via Autolab)
M 9/30	K-NN, Decision Trees, and Rule Learning	William
W 10/2	Evaluating and Comparing Classifiers Experimentally	William	HW5: Compare classifiers Download: data1 data2 (due Oct. 9th (Before lecture) via Autolab) Example code: [1]
M 10/7	PAC Learning	Eric (William out)
W 10/9	Bias-Variance Decomposition	Eric (William out)	HW6: PAC and VC dimension
M 10/14	Ensemble Methods 1	William
W 10/16	Ensemble Methods 2	William	Project description: [2] and Project Milestone 1
M 10/21	Unsupervised Learning: k-Means and Mixtures	Eric
W 10/23	Unsupervised Learning: Dimensionality Reduction	Eric	Project milestone 2: Description, Classifier assignments, and Datasets for Milestone 2
M 10/28	Semi-Supervised Learning	William
W 10/30	Collaborative Filtering and Matrix Factorization	William	Project milestone 3 Milestone3 Handout
M 11/4	Graphical Models 1	Eric
W 11/6	Graphical Models 2	Eric	HW7: Graphical Models (due Nov. 13th before class via BlackBoard)
M 11/11	HMMS, Sequences, and Structured Output Prediction	William
W 11/13	d-separation, Explaining away, and Topic Models	William	Project milestone 4
M 11/18	Network Models	Eric
W 11/20	Review Session/Special Topics	Eric
M 11/25	Not-quite-final Exam	ExamSolutions
W 11/27	No class - Thanksgiving
M 12/2	Markov Decision Processes and Reinforcement Learning	Eric
W 12/4	Scalable Learning and Parallelization	William	Milestones 5-6 Description
Mon 12/9			Milestone 5 due
Tue 12/10			Milestone 6 (writeup) due

To other instructors: if you'd like to use any of the materials found here, you're absolutely welcome to do so, but please acknowledge their ultimate source somewhere.

Section-by-Section

Linear Classifiers

A probabilistic view of linear classification:

• 10-601 Introduction to Probability
• 10-601 Naive Bayes
• 10-601 Logistic Regression

Another view of classification:

• 10-601 Introduction to Linear Algebra
• 10-601 Perceptrons and Voted Perceptrons
• 10-601 Voted Perceptrons and Support Vector Machines

Summary:

• 10-601 Wrap-up on Linear Classification