Comp 150-CLT: Computational Learning Theory
Department of Computer Science
Tufts University
Spring 2004

Course Web Page (this page): http://www.eecs.tufts.edu/comp/150CLT/

Syllabus:

Description: Can machines learn? And what exactly do we mean by this question? This course is concerned with formal models of machine learning, the computational problems associated with such models, their feasibility and complexity, and efficient algorithms for them. Several aspects and models will be studied including: learning as on-line prediction, learning in probabilistic settings, learning from noisy (corrupted) data, and learning by asking questions. For each of these we will have well defined criteria of A successful learning, and study the algorithmic questions: can this be done? And what are good algorithms for the task? Topics will include foundational results as well as recent developments and their impact in applications thereby providing a good grounding in the field.

Prerequisites:

Math 22 and Comp 160 or similar background. Some knowledge of probability theory (e.g. from math 161). Comp 170 is also helpful but not required. the course requires an aptitude for mathematical analysis, writing proofs etc.; where needed additional relevant background will be introduced.

Class Times:

Tuesday, Thursday 4:00-5:15 Halligan H-106

Instructor:

Roni Khardon
Office: Halligan 230
Office Hours: Tuesday 1:15-2:55, Thursday 9:30-10:50 or by appointment
Phone: 1-617-627-5290
Fax: 1-617-627-3220
Dept.: 1-617-627-3217
Email: roni@cs.tufts.edu

Teaching Assistant:

Julie Weber
Email: jweber@cs.tufts.edu
Office Hours: Tue, Thu 3-4 in Halligan 127 (Conference room)

Course Work and Marking

The course mark will be determined as follows:

Collaboration:

Please consult the University guidelines on plagiarism for general information. The problems assigned in this course typically require a good amount of thought and typically cannot be done in one sitting. In this process you are allowed to dicsuss ideas and approches with other students and well as the instructor and TA. You should not share complete solutions with other students or obtain such solutions in any other way. Additionally, any such level of exchange of ideas should be clearly stated on the assignment. In any case I expect you to do your own work and write it up yourself.

Textbooks and Material Covered

Much of the material will be taken from the first text and I recommend purchasing it. Some material from the second text will be covered. Other material will be taken from article and notes that will be provided. The other books may give you a slightly different perspective on the same material so it may be useful to have a look. All the texts are on reserve in the library.

Class Handouts

Additional Articles and Pointers

Homework Assignments