The Machine Learning Track
The Machine Learning Track is intended for students who wish to develop their knowledge of machine learning
techniques and applications. Machine learning is a rapidly expanding field with many applications in diverse
areas such as bioinformatics, fraud detection, intelligent systems, perception, finance, information retrieval,
and other areas.
Track Advisor
Prof. Tony Jebara is
the advisor for Masters students following this track. E-mail: jebara@cs.columbia.edu.
Overall Requirements
Students must complete a total of 30 credits:
1. Fulfill the 12-credit core breadth requirement.
2. Two required courses (6 credits): select from among COMS W4771, COMS W4252, and COMS W6772.
3. Six (6) elective credits at the 6000-level. At least 3 of these credits must be selected from the Electives list below.
4. Six (6) credits of general elective graduate courses at 4000-level or above; at least 3 of these
credits must be CS graduate courses.
5. Students using previous courses to fulfill core or track requirements may complete the 30 graduate credits
by expanding their electives selected from (a) the list of required track courses; (b) the list of elective
track courses; or (c) other graduate courses. At most 3 credits overall may be from non-technical
graduate courses.
Core Breadth Requirements:
Students must complete at least four Core courses out of the following six:
COMS W4115: Programming Languages and Translators
COMS W4118: Operating Systems
COMS W4156: Advanced Software Engineering
CSOR W4231: Analysis of Algorithms
COMS W4701: Artificial Intelligence
CSEE W4824: Computer Architecture
Required Track Courses
Students are required to complete two (2) of the following courses:
COMS W4771: Machine Learning
COMS W4252: Introduction to Computational Learning Theory
COMS W6772: Advanced Machine Learning
Students who have completed equivalent courses with grades of at least 3.0 may apply those courses to
satisfy these requirements and devote more credits to pursue elective courses.
Elective Track Courses
Students are required to complete two courses (6 credits) from the following list; at least one course
must be a 6000-level course. Other courses on this list may be used as general or to replace core or
required track courses when the student has received a waiver.
COMS W4111: Database Systems
COMS W4252: Introduction to Computational Learning Theory
COMS W4731: Computer Vision
COMS W4705: Introduction to Natural Language Processing
CBMF W4761: Computational Genomics
COMS W4771: Machine Learning
COMS E6111: Advanced Database Systems
COMS E6253: Advanced Topics in Computational Learning Theory
COMS W6771: Advanced Machine Learning
COMS E6901: Projects in Computer Science
COMS E6998: Advanced Learning Theory
ELEN E6717: Information Theory
IEOR E4007: Optimization Models and Methods
SIEO W4150: Introduction to Probability and Statistics/
General Electives
Candidates are required to complete at least 6 additional graduate credits at, or above, the 4000 level; at
least 3 of these credits must be CS, the other 3 credits may be a technical or non-technical elective approved
by the track advisor. At most 3 credits overall of the 30 graduate credits required for the MS degree may be
non-technical.
*Note: The list of electives may be updated to reflect changes in the schedule of course offerings.