CS 450 Numerical Analysis

Modified Grading Scale

This is the modified grading scale used for overall course letter grades. The scale is the same for both graduate and undergraduate students.

Course staff

Instructor

Teaching assistants

Textbook

Scientific Computing: An Introductory Survey by Michael T. Heath, McGraw-Hill, 2nd edition, 2002

Discussion forum

https://piazza.com/illinois/fall2016/cs450/home

Video recordings of lectures

https://recordings.engineering.illinois.edu:8443/ess/portal/section/c24309fe-6205-44fb-808c-e6a4819d65a9

Exams, quizzes, and homework

Exams: one midterm and a final exam, both of which will be offered in the Computer-Based Testing Facility (located in room 57 Grainger Library).

• Midterm exam: Tuesday, October 4, 2016 - Friday, October 7, 2016
• Final exam: Thursday, December 8, 2016 - Tuesday, December 13, 2016

Quizzes: approximately weekly, to be taken online on this website (see detailed schedule below). Homework: assignments due every two weeks (see detailed schedule below). Homework will involve both written exercises and computer problems. The latter must be done in Python. Projects: students taking the course for 4 credit hours must complete a term project. See detailed schedule below for various due dates. More project details can be found at the bottom of this page.

More detailed policies on exams, quizzes, and homework are given below.

Course Policies

Homework Policies:

• Homeworks have one graded attempt. Once your assignment is submitted, you cannot change your solutions. Homework submissions will not be reopened.
• As soon as you start the homework, you should make sure that the desired option for "At deadline" is selected. If "Submit session for grading is selected", your assignment will be automatically submitted at the deadline. If you wish to work on the assignment after the deadline and submit late, you must select "Do not submit session for grading". Otherwise, your assignment will automatically be submitted at the deadline and your session will not be reopened.
• Late Policy: Homeworks are due at 5pm on the due date. Homework submitted after the deadline will receive half-credit (i.e., your work will be graded and multiplied by 0.5). You can submit homework for half-credit for up to one week. You must "submit assignment" and "confirm" before the deadline; if you start a homework before the deadline, but finish after the deadline, then you will receive half-credit. No credit will be awarded for homework more than one week late.
• Solutions to the homework are viewable (and reviewable) after the due date. If you are submitting homework for the half-credit due date, you can look at the solutions, but you cannot directly copy the sample solution. What you submit must still be written by you, not just copied.
• A homework is considered either all late or all on-time. If any portion of the homework is submitted after the deadline then the entire homework will be graded as late.
• The work you submit in your homework must be your own. Collaboration among students is not permitted. General discussion among students is acceptable, but detailed sharing of code or solutions is prohibited.
• For any homework problems requiring file submissions, the uploaded file must be in PDF format, clearly legible, with all problem parts clearly labeled. For non-coding problems with file upload submissions, you must show your work (showing how you arrived at your answer and/or explaining your answer as appropriate). For these problems, an answer without appropriate work will receive little or no credit.
• For coding problems, it is your responsibility to ensure that the code runs without errors and produces the desired output on Relate. The course staff will not change your code or reopen your homework submission to correct mistakes.
• All plots should be clearly titled, labeled, and have legends as appropriate. Failure to do so will result in loss of points.

Quiz Policies:

• Quizzes have one graded attempt. Once you submit a quiz, you cannot change your answers.
• Quizzes are due 10 minutes before the start of class on the day they are due (see schedule below).
• Unlike homework, you must fully submit your quiz (there is no automatic submission at the deadline). If you do not submit your quiz (by pressing "Submit assigment" in the upper right and confirming submission), you will receive no credit for the quiz (regardless of whether or not answers were saved for individual questions).
• No late quiz submissions will be accepted.

Piazza Policies:

• Piazza will be used for important class communications. It is your responsibility to monitor Piazza for Instructor notes.
• The course staff will not debug source code via Piazza. If you would like help debugging your code, you will need to come to office hours.
• It's okay to ask questions regarding homework problems, but please do not post your solutions (e.g. large blocks of code, handwritten solutions, etc).
• Before posting a question, search for keywords from your question. This helps minimize duplicate questions, making it easier for other students to search for answers to their questions.
• Please ensure that your name on Piazza is set to your actual name.

Exam Policies:

• All exams must be taken in the CBTF during the testing period.
• No exams will be administered outside of the CBTF.
• Because we allow a multiple day period for you to schedule your exam, no makeup or conflict exams will be given except in cases of severe emergencies. We recommend not signing up for last day, particularly if you are likely to get sick.

Class schedule

|date|topic|homework|quiz|project (4-credit only)| |---|---|---|---|---|---| |Mon Aug 22|Scientific computing| | | | |Wed Aug 24|Scientific computing|HW1 assigned|Quiz 0 due (no credit quiz)| | |Mon Aug 29|Linear systems| |Quiz 1 due| | |Wed Aug 31|Linear systems| | | | |Mon Sep 5|Labor Day holiday| | | | |Wed Sep 7|Linear systems|HW1 due, HW2 assigned| | | |Mon Sep 12|Linear least squares| | Quiz 2 due| | |Wed Sep 14|Linear least squares|example| | | |Mon Sep 19|Eigenvalue problems| | Quiz 3 due| | |Wed Sep 21|Eigenvalue problems|HW2 due, HW3 assigned| | | |Mon Sep 26|Eigenvalue problems| | | | |Wed Sep 28|Nonlinear equations| | Quiz 4 due| | |Mon 0ct 3|Nonlinear equations| | | | |Wed 0ct 5|Optimization|HW3 due, HW4 assigned| Quiz 5 due| | |Mon 0ct 10|Optimization| | | Proposal due| |Wed 0ct 12|Optimization| | | | |Mon 0ct 17|Interpolation| | Quiz 6 due| | |Wed 0ct 19|Interpolation|HW4 due, HW5 assigned| | | |Mon 0ct 24|Numerical quadrature| | Quiz 7 due| | |Wed 0ct 26|Numerical quadrature| | | | |Mon 0ct 31|IVPs for ODEs| | Quiz 8 due| | |Wed Nov 2|IVPs for ODEs|HW5 due, HW6 assigned| | | |Mon Nov 7|IVPs for ODEs| | | Progress report due| |Wed Nov 9|BVPs for ODEs| | Quiz 9 due| | |Mon Nov 14|BVPs for ODEs| | | | |Wed Nov 16|PDEs|HW6 due, HW7 assigned| Quiz 10 due| | |Mon Nov 21|Fall break| | | | |Wed Nov 23|Fall break| | | | |Mon Nov 28|PDEs| | | | |Wed Nov 30|PDEs| | | | |Mon Dec 5|FFT| | Quiz 11 due| | |Wed Dec 7|Stochastic simulation|HW7 due| |Final report due|

Grading Information

Grade Breakdown

Grade Scale

Academic Grading System in the US

4 Credit Hour Project Details

All students taking the course for 4 credit hours must complete a project. This project is chosen by the student with the consent of the instructor.

To ensure that a given project is appropriate in scope and content, students must submit a brief description of the proposed project for approval before beginning implementation. The instructor may suggest modifications or alternatives, if appropriate. Projects are evaluated for both correctness and creativity. Projects are graded as satisfactory/unsatisfactory and make no contribution to the final grade, only to the hours of credit received.

Progress proposals, reports and final reports will be submitted on Relate (the course website). Additional details regarding expectations for the progress report and the final project report will be given in the individual submission spots. Due dates for each portion are shown in the schedule above.

Possible types of projects include:

• An education module that illustrates or demonstrates an important concept or algorithm covered in the course.
• A research project on some general numerical method.
• A research project in which a student applies a numerical method to solve a specific problem arising in the student's own research area.
• If none of the previous categories is appealing, then a project consisting of extra homework (assigned by the instructor) is acceptable. In this case, the project proposal should specify a preferred topic (e.g. textbook chapter) on which the homework will be chosen.

If you are unsure about what project to propose or what sort of project is acceptable, feel free to consult the instructor.