|
RIT Department of Computer Science
Pattern Recognition (Topics in Artificial Intelligence) Fall 2010 | 
|
|
| 4005-759-01 (Lecture Room: 70-3500 (DPRL lab)), Instructor: Richard Zanibbi | ||
| Office hours: 10-11:00am Tues. and Thurs., 1-3:00pm Fridays (70-3551, Golisano College) |
Home --- Syllabus --- Slides --- Assignments --- Programming --- Resources --- [ MyCourses ]
An introduction to pattern classification and structural pattern recognition. Topics include: feature extraction, Bayesian decision theory, nearest-neighbor rules, clustering, support vector machines, neural networks, classifier combination, and syntactic pattern recognition techniques such as stochastic context-free grammars. The course is part lecture and part seminar: students will present some course material to the class as well as complete and present a research paper. In addition, programming assignments will provide students with practical experience in constructing pattern recognition systems such as optical character recognizers (OCR). Class hours: 4, Credit: 4
Prerequisites: 4003-455 (Artificial Intelligence) or 4005-750 (Introduction to Artificial Intelligence), or Permission of the instructor.
At the end of this course, students will be able to:
Recommended Text [ Web Page ]: Combining Pattern Classifiers: Methods and Algorithms, by Lumila I. Kuncheva, Wiley 2004
Course Web Page: http://www.cs.rit.edu/~rlaz/2010/
We will be working from multiple sources this quarter, and a number of readings will be posted on the MyCourses page for the course.
There is a significant amount of mathematical content in the course. If you find that you are challenged by notation or mathematical concepts used in the course, please make use of the instructor's office hours, or email the instructor to set up an appointment. If you need additional help after consulting the instructor, you are encouraged to make use of the RIT Academic Support Center, which has a drop-in center and a number of other useful resources.
The course is part seminar; students will present some course material in pairs. Individually, students will complete programming assignments and a research survey on a pattern recognition topic.
| 35% | Programming assignments (2) (10%, 25%) |
| 30% | Seminar presentations (3) |
| (20 mins., plus 5 mins. for questions) | |
| 30% | Research Paper |
| (Proposal: 5%, Presentation (10 mins. + plus 2 mins. questions): 5%, Final Paper (12-15 pages): 20%) | |
| 5% | Class Participation |
Important Note: this is an advanced graduate course in the Intelligent Systems cluster. The following are expected and taken into consideration when grading:
Students will give three presentations in groups of two over the quarter, as assigned by the instructor. The materials to present will be assigned, but groups will be expected to produce slides and/or other notes for the class to use. These materials will be posted on the course web pages.
Presentations will be given in weeks 4, 6, and 8. Topics will be assigned two weeks before a presentation is to be given.
Each seminar has a weight of 10% of the final grade, with 5% for the 20-minute presentation given in-class (+5 minutes of questions), and 5% for slides and other materials provided.
Research papers by students that previously took the course are available here.
Students will submit a 3-4 page topic proposal (including at least 3 relevant research papers) in Week 5. The instructor will provide feedback on the topic and chosen references. Students are strongly encouraged to submit a draft of their paper by Tuesday of Week 9 to the instructor, in order to feedback on the paper itself. No grade is associated with the draft.
Students are strongly encouraged to find a survey paper on their research topic, as this can save tremendous amounts of time.
The final paper should be 12-15 pages in length, and include a minimum of 10 references. Expectations regarding style, organization and content are summarized here. In week 10, research paper presentations will be given in-class (10 mins. + 2 mins. questions), providing another opportunity for feedback before the final paper is submitted.
Formatting: Topic proposals and research papers will be produced using LaTeX (which is widely used in Computer Science research), making use of the IEEE conference paper style file available here. The formatting should be reasonably dense, i.e. appear the way papers do in regular IEEE conference proceedings.
Proposals and papers should be submitted through MyCourses as .pdf files.
Please note that this schedule may change over the course of the quarter:
| Week | Topics | Assignments and Exams
| 1 | Overview, Bayes' Decision Theory |
| 2 | Linear and k-Nearest Neighbor Classifiers |
| 3 | Decision Trees | Thurs: Assign 1 due
| 4 | Neural Networks | [Student presentations]
| 5 | Support Vector Machines | Thurs: Paper proposal due
| 6 | Classifier Combination | [Student presentations]
| 7 | Feature Selection | Thurs: Assign 2 due
| 8 | Clustering | [Student Presentations] | 9 | Structural and Syntactic Pattern Recognition |
Tues: Paper draft
| 10 | Review, Paper Presentations | Tues: Assign 3 due | Thurs: Research paper presentations 11 | (Exam week) | Friday, 9am: Final Paper Due
| |
If you have special needs for seating, tests, note-taking services or other matters due to a disability, please contact the Disability Services Office (www.rit.edu/dso). If you receive approval for accomodation within the course, please contact me as soon as possible so that we can make the necessary arrangements.
Late submissions may be submitted at most two days late, with a 10% grade penalty each day. After two days (11:59pm on the second day after the deadline), late submissions will not be accepted.
Exams will only be rescheduled in the case of difficult situations for which there is formal documentation (e.g. a doctor's note). Contact the instructor as soon as possible if you encounter scheduling or other issues regarding the exams.
Students may discuss assignments and projects with others, but submitted work (papers and code) must be created independently by each student or group, and not copied from another student or other source.
All borrowed ideas, text, or code used in papers and assignments must be cited appropriately. Citations must be provided using an accepted format for academic journals in computer science (e.g. ACM or IEEE styles). Providing only a URL for a references is unacceptable: in the case where a URL is appropriate (e.g. for software), the author, title, and date for the document associated with the link must be provided with the citation. In cases where it appears that copying of material or plagiarism has occurred, the Department of Computer Science Policy on Academic Integrity will be followed.