Final Project Reminders: The exam (project presentations) will be given Friday Dec. 9th from 1:30-4pm in our regular classroom. Presentation slides are due 1 hour before the exam through MyCourses (the dropbox will close at 12:30). The final project report and code are due the following Monday (Dec. 12th, by 11:59pm). Submission links and the concrete evaluation rubrics for the presentation, report, are also available in MyCourses.
**Assignment 5 is now optional; if you complete it, it will replace your lowest assignment grade. It will be released by Tuesday of next week, and will be due on Wednesday Dec. 14th (after the proposal is due) at 11:59pm. No late submissions will be accepted.
Quiz 10 will be released Thursday evening, and will be due on Friday.
Readings: Chs. 4 and 5 -- see MyCourses
Course Evaluations are now open, and will close before exams begin (not sure whether this is Monday of Wk 16, or the Reading Day on Tuesday). I value your feedback! Please take a moment to share what you think worked in the course, and what could be improved for future students.
Week 14
**No RIT classes Wed-Friday this week. Enjoy the holiday break!
**Make-up Assignments: Students are permitted to redo any of Assignments 1-3 if re-submitted through MyCourses by this Tuesday (Nov. 22nd) at 11:59pm. 20% late penalty applies.
Quiz 9 will be released by Monday morning, and will be due Tuesday morning at 1:00pm.
Readings: Ch. 4 -- see MyCourses
Course Evaluations are now open, and will close before exams begin (not sure whether this is Monday of Wk 16, or the Reading Day on Tuesday). I value your feedback! Please take a moment to share what you think worked in the course, and what could be improved for future students.
Week 13
**Office Hours are cancelled on Tuesday. Please send email or messages over discord if you have questions about the assignment or anything else.
**Syllabus update: There will be only 5 assignments this term rather than 6 (60% of your final grade). As a result, each assignment will now be worth 12% of your final grade.
Quiz 8 was due Tuesday at 9pm.
Assignment 4 is due on Thursday (11:59pm). For this assignment, late submission will be accepted until next Tuesday (Nov. 22 -- 20% penalty applies).
Readings: Ch. 4 (Recurrent NNs, and Word Embeddings) -- see MyCourses
Project: We will discuss the course project proposals this week (Thursday)
Make-up Assignments: Students are permitted to redo any of Assignments 1-3 if re-submitted through MyCourses by Tuesday, Nov. 22nd at 11:59pm (before the Thanksgiving break). 20% late penalty applies.
Course Evaluations are now open, and will close before exams begin (not sure whether this is Monday of Wk 16, or the Reading Day on Tuesday). I value your feedback! Please take a moment to complete the course evaluation, and let me know what you think has worked in the course, and what might be improved for future students.
Week 12
Readings: Charniak Ch. 3 and Ch. 4 (Recurrent NNs, and Word Embeddings) -- see MyCourses
**Make-up Assignments: Students are permitted to redo any of Assignments 1-3 if re-submitted through MyCourses by Tuesday, Nov. 22nd at 11:59pm (before the Thanksgiving break). Note that the 20% late penalty will be applied to any resubmitted assignment (i.e., the maximum grade for any one assignment is 80%)
Assignment 4 will be released on Thursday, and will be due Tuesday of Wk 13.
Quiz 8 will be released Friday, and will be due at 1pm on Monday of Wk 13.
Week 11
**Extension: Project proposal is now due Friday at 11:59pm
**Make-up Assignments: Students are permitted to redo any of Assignments 1-3 if re-submitted through MyCourses by Tuesday, Nov. 22nd at 11:59pm (before the Thanksgiving break). Note that the 20% late penalty will be applied to any resubmitted assignment (i.e., the maximum grade for any one assignment is 80%).
Readings: Charniak Chs. 3 and 4 (Recurrent NNs, and Word Embeddings)
**Preparing for upcoming assignment and the project -- make sure to run and experiment with the code examples from Charniak Chs 2 and 3. The single linear layer + softmax MNIST digit recognizer shown in Fig 2.2 is provided in a (modified) program was already shared on MyCourses. To solidify understanding, students should manually copy and run the example programs and/or modify the provided Fig 2.2 code. (Hint: keeping models separate will allow you to experiment more easily with the performance and speed of different models)
Assignment 3 grades are out -- make sure to review comments on your submission and the general comments to the class carefully. If you have questions, contact Prof. Zanibbi via email.
Week 10
Prof. Zanibbi's office hours on Thursday are cancelled. Please do send email if you have questions, and I will respond as soon as I am able to.
New TA Office Hours: Ayush will now be holding office hours Tuesdays and Wednesdays 12pm-1:50pm. For this Wednesday, he will be available over Zoom, and starting next week will also be available in-person in the Cybersecurity building (CYB-2795, which should be the same room that was being used previously for the Tuesday TA meeting space).
Readings: Charniak Ch. 3 (Convolutional NNs)
Group Project: Group requests are due Wednesday; Thursday any remaining students will be assigned to groups by the instructor. The Project Proposal is due next Thursday at 11:59pm (Nov. 3rd) through the MyCourses assignment 'drop box.' See the project write-up for requirements, and the recommended approach to completing the course project.
Assignment 4 will be released by early next week (Week 11).
There will be no Quiz this week.
Week 9
Readings: Charniak DL Chs 1 & 2 (same as last week)
Deliverables:
A *draft* for the group project requirements has been posted on MyCourses. We will discuss the project in-class on Thursday, after which the project requirements will be finalized.
Groups will have 3 students, and must be chosen by students Oct. 26 (next Wednesday), after which the instructor will choose groups for unassigned students.
Project proposal is due Thursday Nov. 3rd.
Assignment 3 is due Tuesday at 11:59pm.
Quiz 7 will be released Wednesday, and be due Thursday at 1pm (1 hr before class).
Assignment 4 will be released later this week, and will be due next Thursday.
Week 8
**No Lecture on Tuesday** (RIT Holiday)
**No Quiz** this week due to the holiday.
Exam date:Friday, Dec. 9th from 1:30-4:00pm (in our regular classroom, CBT 1160)
Readings: Charniak DL Ch 1, Ch 2
Re: Reading; remember this saying attributed to Michael Kalish, "A year in the lab saves you a day in the library." Reading and study build understanding that you can use to organize, diagnose, and explain program behavior (e.g., 'what is this program intended to compute exactly, and how?'). Strong applied CS work requires substantial amounts of both reading/study and lab work. However, in terms of efficient implementation and analysis of complex programs, and in creating innovative computational models and implementations, reading and study generally get you (much) further.
Assignment 3 will be released early this week, and will be due Tuesday Oct. 18 at 11:59pm.
Week 7
Exam date:Friday, Dec. 9th from 1:30-4:00pm (in our regular classroom, CBT 1160)
Readings: Hastie 9.2, Hastie Ch. 15 (skim for details of interest), Criminisi et. al Ch 3.1-3.4 on Random Forests, similarities of margin properties of SVMs (see MyCourses), Charniak DL Ch 1
Assignment 2: is due Tuesday at 11:59pm.
Quiz 6: will be released Wednesday morning, and be due Thursday at 1pm.
Assignment 3 will be released later this week, and be due at the end of next week.
Schedule Reminder: No classes on Oct 10/11 next week (October break), no lecture next Tuesday
Week 6
Assignment 2: will be released on Tuesday, and will be due next Tuesday at 11:59pm.
Quiz 5: was released Wednesday evening, and is due Friday morning at 10am.
Week 5
Quiz 4 will be released by Wednesday at noon, and will be due Thursday at 1pm.
Readings: Hastie 12.1 - 12.3.1 (SVMs)
Week 4
This week we will continue our discussion of Bayesian Decision Theory. Associated readings may be found in MyCourses (Duin Ch 2)
Quiz 3 will be released by Wednesday at noon, and will be due Thursday at 1pm.
Week 3
This week we will discuss probability and Bayesian Decision Theory. Associated readings may be found in MyCourses (Charniak Ch 2, and Duin Ch 2)
Quiz 2 will be released by Wednesday at noon, and will be due Thursday at 1pm.
Assignment 1 is due Tuesday, Sept. 6th at 11:59. Make sure to follow the provided instructions to install and run the program, and use the 'prepsubmit' program before submitting your code and source code listing (as code.zip and listing.pdf).
Lectures this week: Prof. Zanibbi is away; Tuesday's lecture will be recorded and posted to MyCourses by Tuesday afternoon. Thursday's lecture will be given by Ayush (our TA).
Week 2
This week we will discuss parametric and non-parametric classification, looking at a least-squares linear classifier, and k-nearest neighbor as examples. Please read Hastie 2.1-2.3 (as indicated in the Schedule)
Quiz 1 will be released by Wednesday at noon, and will be due Thursday at 1pm.
Assignment 1 will be released through MyCourses on Tuesday evening, and will be due next Tuesday, Sept. 6th at 11:59 (submission through MyCourses). Make sure to follow the provided instructions to install and run the program, and use the 'prepsubmit' program before submitting your code and source code listing (as code.zip and listing.pdf). Information regarding which machine to use in ICL6, and expected programming style are available on MyCourses under Programming , along with a numpy reference card and documentation.
Lectures next week: Prof. Zanibbi is away next week. Tuesday's lecture will be recorded and posted to MyCourses by Tuesday afternoon. Thursday's lecture will be given by Ayush (our TA).
Week 1
Welcome to the web pages for the machine learning course being offered by the Department of Computer Science at RIT in Spring 2022.
Readings and example applications for Week 1 are listed on the Schedule page. Students in the course can sign up for the class discord server through MyCourses (available now).
Prof. Zanibbi is the course instructor. Ayush Kumar Shah is the TA for the course.
These web pages will be used to communicate information about the course, along with news, deadlines, etc. Assignments and projects will be distributed and submitted using MyCourses.
Lectures: Tuesdays and Thursdays, 2:00pm - 3:15pm Biosciences Building (CBT) Rm. 1160
Due to the ongoing COVID situation, lectures will both be in-person and over Zoom, and lectures will be recorded. Lecture attendance is strongly recommended - students may be tested on additional points discussed in lecture.
For those interested in current machine learning research, a list of conferences and journals in machine learning and some application areas is provided on the Resources page.