George Mason University
School of Information Technology and Engineering
Department of Computer Science
CS 782 Machine Learning
Meeting time: Thursday 7:20pm – 10pm
Meeting location: ST-II, Room 430A
Instructor: Dr. Gheorghe Tecuci, Professor of Computer Science
Office hours: Monday and Thursday, 6pm-7pm, for both CS580 and CS782
Office: ST-II, Room 421
Phone: 703 993 1722
E-mail: tecuci@gmu.edu
Course Description
Machine Learning is concerned with developing intelligent adaptive systems that are able to improve their competence and/or efficiency through learning from input data, from a human user, or from their own problem solving experience. This course presents the principles, strategies, major methods, systems, open issues and research directions in Machine Learning, preparing the students to build evolving knowledge-based systems. Covered topics include: foundations of machine learning, rote learning, empirical inductive learning from examples, explanation-based learning, abductive learning, instance-based learning, case-based reasoning and learning by analogy. The relative strengths and weaknesses of these strategies, as well as their most appropriate application domains will be discussed.
The course will also cover multistrategy learning which is concerned with building advanced learning systems that integrate several basic learning strategies in a synergistic way, in order to solve learning tasks that are beyond the capabilities of the integrated strategies. It will also discuss automated knowledge acquisition, integrated teaching and learning, and instructable agents, as well as open issues, current trends and frontier research in Machine Learning.
Interested students may work on projects relevant to their research area or may experiment with or enhance the Disciple learning and problem solving agent developed in the Learning Agents Laboratory.
This course can be used as a Ph.D. comprehensive course.
Grading Policy
There will be assignments, a mid-term exam, and a final exam. It will be permissible, on an individual basis, to replace the final exam with a project. This option, which may be of high interest to PhD students, requires an early submission of a proposal.
The course grade will be determined as follows:
Assignments 5%
Mid-term exam 50%
Final exam or project 40%
Class participation 5%
Extra credit:
Assignment3 10%
Readings
Lecture notes provided by the instructor (required).
Tom Mitchell, Machine Learning, New York: McGraw Hill, 1997 (recommended).
Additional relevant references
Buchanan, B. G. and Wilkins, D. C. (editors), (1993). Readings in Knowledge Acquisition and Learning: Automating the Construction and Improvement of Expert Systems, Morgan Kaufmann, San Mateo, CA.
Shavlik, J. W. and Dietterich, T. (editors), (1990). Readings in Machine Learning, Morgan Kaufmann.
Tecuci, G. (1998), Building Intelligent Agents: An Apprenticeship Multistrategy Learning Approach, Academic Press.
Michalski, R. S. and Tecuci, G. (editors), (1994). Machine Learning: A Multistrategy Approach, Volume 4, Morgan Kaufmann Publishers, San Mateo, CA.
Tecuci, G. and Kodratoff, Y. (editors), (1995). Machine Learning and Knowledge Acquisition: Integrated Approaches, Academic Press.
Exams and Assignments
10/16/2003 Midterm Exam (new date)
Assignment 1 (due date: 10/9/2003)
Assignment 2 (due date: 10/16/2003)
Assignment 3 (due date: anytime before 11/20/2003)
Lecture Notes
Inductive Learning from Examples: Version Space Learning
Inductive Learning from Examples: Decision Tree Learning
Evaluation of Inductive Learners
Case-based Reasoning and Learning
Multistrategy Learning
Presentation
Paper: An inference-based framework for multistrategy learning
Learning and Problem Solving Agents