Machine Learning (CSL 407) Fall 2014

Timings and Lecture Hall:

Lecture Hall - L2

Class Lectures - Wednesday (4.15-5.05pm), Thursday (1.30-2.20pm) and Friday (2.25-3.15pm)

Lab hours - Monday (1.30-3.15pm) and Thursday (9.00-10.45am) - Lab hours are not mandatory. However the students are advised to take advantage of these hours to talk to the TA for any assistance with the course.

Description:This course provides a detailed investigation of current machine learning theory and methodologies, along with hands on experience on machine learning algorithms through programming in Matlab.

Prerequisites: Data Structures (CSL 201)

Textbook(s): There is no fixed textbook for the course. However content will be adopted from the following textbooks

Machine Learning by Tom Mitchell (ML)
Elements of Statistical Learning by Trevor Hastie, Robert Tibshirani and Jerome Friedman (available online for free) (ESL)
Pattern Recognition and Machine Learning by Christopher Bishop (PRML)

Grading: Homeworks-6 out of 7(30%), Pop Quizzes (5%), Presentation(10%), Exams(50%), Class Participation (5%). A student has to secure an overall score of 40 (out of 100) and a combined score of 60 (out of 200) in the exams to pass the course.

Course Details: PDF

Instructor: Narayanan C Krishnan (CK)

Office Hours: Wednesday and Thursday 10.00-11.00am or by appointment.

Office: 318

Phone: +91 1881 242273

Email: ckn@iitrpr.ac.in

Teaching Assistant: Yayati Gupta and Akrati Saxena

Office Hours: Monday (1.30-3.15pm) and Thursday (9.00-10.45am) or by appointment

Office: 120

Email: yayati.gupta@iitrpr.ac.in, akrati.saxena@iitrpr.ac.in

Grades: PDF

Homework:

Homework 1 is due on 15 Aug 2014 11.59pm.
Homework 2 is due on 29 Aug 2014 11.59pm.
Homework 3 is due on 12 Sep 2014 11.55pm.
Homework 4 is due on 17 Oct 2014 11.55pm.
Homework 5 is due on 29 Oct 2014 11.55pm.
Homework 6 is due on 14 Nov 2014 11.55pm.
Homework 7 is due on 21 Nov 2014 11.55pm.

Presentations: Guidelines, Time: Tuesday 4.00-5.00pm L3

Available slots are colored in black

Group 1 and 2 - Oct 7

Group 3 and 4 - Oct 14

Group 5 and 6 - Oct 28

Group 7 and 8 - Nov 4

Group 9 and 10 - Nov 11

Group 11 and 12 - Nov 18

Group 13 and 14 - Nov 19 (regular class hour)

Group 15 and 16 - Nov 20 (regular class hour)

Group 17 and 18 - Nov 21 (regular class hour)

Course Schedule (Tentative)- Lectures* and Deadlines

Week	Date	Topic	Readings	Submission Deadlines
1	July 30	Introduction
	July 31	Linear Models of Regression	PRML (1.5.5, 3.1,3.2) Linear Regression Notes by Andrew Ng
	Aug 1	Linear Models of Regression	PRML Appendix A-E
2	Aug 6	Linear Models of Regression
	Aug 7	Linear Models of Classification	ESL (4.1, 4.2)
	Aug 8	Linear Discriminants	ESL (4.3)
3	Aug 13	Logistic Regression	PRML (4.3), ESL (4.4) Logistic Regression Notes by Andrew Ng
	Aug 14	Logistic Regression
	Aug 15	Independence Day (holiday)		HW1
4	Aug 20	Perceptron	ML (4) PRML (4.1.7, 5.5.1, 5.5.2)
	Aug 21	No class
	Aug 22	Multi-Layer Perceptron
5	Aug 27	Multi-Layer Perceptron
	Aug 28	Support Vector Machines	Slides from Andrew Ng's lecture Part 1
	Aug 29	Support Vector Machines	Part 2	HW2
6	Sep 3	Support Vector Machines
	Sep 4	Decision Tree Learning	ML (3)
	Sep 5	Decision Tree Learning
7	Sep 10	Instance Based Learning	ML (8)
	Sep 11	Instance Based Learning
	Sep 12	Instance Based Learning		HW3
8	Sep 17	Performance Measures	ML (5)
	Sep 18	Experimental Design
	Sep 19	Experimental Design
9	Sep 24	Mid Semester Exams
	Sep 25	Mid Semester Exams
	Sep 26	Mid Semester Exams
10	Oct 1	Ensemble Methods	Notes by Tom Dietterich
	Oct 2	Navrathri (Holiday)
	Oct 3	Navrathri (Holiday)
11	Oct 8	Ensemble Methods	Introduction to Boosting by Freund and Schapire PRML (14.3)
	Oct 9	Ensemble Methods	Slides from Nando de Freitas's lecture
	Oct 10	Zeitgeist (Holiday)
12	Oct 15	Clustering	ESL (14.1, 14.3), PRML (9.1)
	Oct 16	Clustering	ESL (14.3), PRML (9.2)
	Oct 17	Clustering	ESL (14.3)	HW4
13	Oct 22	Dimensionality Reduction	PRML (12.1)
	Oct 23	Diwali (holiday)
	Oct 24	Dimensionality Reduction
14	Oct 29	Manifold Learning	Original ISOMAP paper	HW5
	Oct 30	Feature Selection	Feature Selection
	Oct 31	Probabilistic Models	ML (6)
15	Nov 5	Hidden Markov Models	Tutorial on HMM by Rabiner
	Nov 6	Guru Nanak's Birthday (Holiday)
	Nov 7	HIdden Markov Models
16	Nov 12	Hidden Markov Models
	Nov 13	Reinforcement Learning
	Nov 14	Reinforcement Learning		HW6
17	Nov 19	Presentation
	Nov 20	Presentation
	Nov 21	Presentation		HW7
18	Nov 26	End Semester Exams
	Nov 27	End Semester Exams
	Nov 28	End Semester Exams

* Acknowledgement: We are very grateful to Alex Smola, Andrew Moore, Andrew Ng, Chuck Anderson, Ethem Alpaydin, Larry Holder, Nando de Freitas, Pedros Domingos, Ryan Tibshirani, Russell Greiner, Thorsten Joachims, Tom Mitchell for allowing us to use some of their course material as part of this course.

Resources

Machine Learning