AMATH 352 Spring 2016


Applied Linear Algebra and 

Numerical Analysis


Instructor: Niall Mangan, niallmm at uw dot edu
Office hours: Lewis 116, Mon 4:30-5:30pm, Thurs 9:30-10:30am, Friday 1:30-2:30pm
Teaching Assistant: Felix Xiaofeng Ye, yexf308 at uw dot edu
Office hours: Lewis 115 Tuesday 3-5pm

 We Meet:  MWF 12:30-1:30 in MLR 301

Announcements and Discussion
  • Solution set for Final exam. Final grades are also posted on Catalyst.
  • Grades for Final exam are posted.
  • Grades and Feedback for PS7 are available.  Problem Set 7 Solutions Statistics without zeros:
  •  Mean43
     Median47
     Mode51
     Min7
     Max 52
     Std. Dev.11.86
  • Office Hours by appointment only Exam week.
  • Grades and Feedback for PS6 are available. Please check your feedback (there were some errors earlier, which have been corrected now).
 Mean45
 Median48
 Mode52
 Min27
 Max 52
 Std. Dev.7.56
  •  Mean
     45    
     Median 48
     Mode 48
     Min8
     Max 52
     Std. Dev.8.99
      There was a typo on the Midterm Review. I have updated the pdf. Midterm Review 
    • Solution set for PS3 is available. In this solution set I gave you more directions on how to submit corrections for 1/2 points back on PS3. The deadline for corrections for 1/2 points has past for PS2, so I am also posting the code for that problem set. 
    • Midterm Review is posted.
    • Feedback on PS3 is available in the dropbox I gave you for PS1.  The stats (including bonus, but not including those who did not turn in homework) were:
       Mean
       46.7
       Median 49
       Mode50
       Min25
       Max 52
       Std. Dev.6.39
    • Partial solution set for PS2 is available. Make up for PS2 for half credit are due Wed 4/27. I have not provided code for the algorithms as it would be too difficult to grade make up solutions.
    • Problem Set 4 is posted.
    • Feedback on PS2 is available in the dropbox I gave you from PS1. The stats (including bonus, but not including those who did not turn in homework) were:
       Mean
       41.31 
       Median 44
       Mode 49.0
       Min 17
       Max 52
       Std. Dev. 8.88

    • Feedback on PS1 is posted. Go to the dropbox where you turned it in and click on the link. The stats for grades on PS1 (including bonus) were:
      Mean55.1
      Median58.0
      Mode61.0
      Min18.0
      Max65.0
      Std. Dev.9.39

    • Problem set 3 is now posted. It will be due at 12:30PM on 04/20/2016.
    • There is a correction to problem 2.3c on PS2.  You should analyze the order of convergence by looking at the plot of error(n+1) vs error(n). .
      Here is the code you can use to generate the plot if you have the error vector:

      figure(3)
      loglog(esave(1:N-1), esave(2:N), 'o')
      xlabel('error at n')
      ylabel('error at n+1')
      title('determining order of convergence')

      Using this, you can analyze the slope and determine order of convergence. The Problem set in the link has also been updated. 
    • Solution set for problem set 1 (I only included the .m files for the last problem): 
    • I made a  video reviewing fixed point iteration, that might help you get started on Problem Set 2. I also made a video going through a solution to problems set 1 problem 2.4 c and d. It's a bit long and there is a long pause in the middle. But I don't have time to do much editing... sorry! 
    • Homework 2 is assigned. Download the materials here.
    • Homework 1 is due today. Make sure your zip file is uploaded! Email me if you are having technical difficulties.
    • Latex template for homework now available! Download zip file with latex file for problem set, a matlab script example, and plot file.
    • Office hours now posted ^ above
    • Problem Set 1 Assigned (see left column)
    • If you have never used Matlab before the intro video, intro to the workspace video, and video on scripts and functions will be helpful for the first homework.
    • Homeworks should be submitted to the Catalyst drop box folder here.
    • Due to the large number of students in the class we will not be answering homework or coding related questions via email. Please visit the forum to discuss or scheduled office hours. 
    • Discussion Board Here. Felix and I will monitor the board and provide more resources when you show you are interested or stuck on particular topics, and other students do not respond within 24 hours. You are also encouraged to discuss topics and problems in study groups. However, you must write up your own solutions. Please do not post solutions on the discussion board.
    Pre-requisites: Math 126 or  Q SCI 293

    Schedule: 
    • Homework Deadlines: Homeworks will be due on Wednesdays a week after they are assigned. I will keep the topics and homework schedule updated here.
    • Exams: 
      • Midterm: Wednesday May 4, 2016; 12:30-1:30 (in class); MLR 301
      • Final: Thursday June 9, 2016; 8:30-10:20AM; MLR 301
    • UW Academic Calendar

    Course Description: Nearly every discipline with a quantitative component including engineering, physical sciences, social sciences, finance, computer graphics, big data, and machine learning rely on linear algebra. Numerical computation greatly enables modeling the data analysis in these fields. In order to utilize linear algebra and computing for problem solving, it is essential to understand how to set up problems (in the linear framework and numerically), determine when well defined solutions exist, write programs and algorithms in MATLAB to solve these problems, evaluate whether the algorithm will find the solution efficiently, and evaluate the accuracy of the computation. For a detailed list of topics please go here. 

    Textbook & Resources: No textbook will be required for this course. We will supply class notes for each class here. 

    My notes will be mostly based on this set of notes from Prof. Nathan Kutz, but I will skip around and deviate.

    Here are a few other freely available texts, if I am directly referencing them I will let you know:
    If you want to purchase a textbook as a reference, I really like the explanations in Burden and Faires Numerical Analysis (9th ed). A cheaper alternative that Prof. Kutz notes are based on is Matthews and Fink (3rd ed), but it is denser. For a very focused quick read, I originally learned Numerical Methods from Peter Turner's Guide to Scientific Computing-- also fairly in-expensive.

    Homework and Computing Resources

    Matlab/Ocatave Resources: You need to obtain a copy of or access to MATLAB. You can either buy a student copy for about $100 at the UW Bookstore, get a $30 total headcount license from the link below (expires 9/1/2016), or use the computer labs/remote access listed below. 
    LaTeX Resources for typesetting

    Homework Policy and Grading

    • To encourage you to go over material you had difficulty with, you have an opportunity to make up half the lost points on homework. Within one week of getting your homework back you should find or write an equivalent homework problem, explain why it is an equivalent problem (or why it has changed), and provide a solution. Your updated solution must show that you now understand the concept(s) covered by the problem, and have not just switched out numbers. 
    • We strongly encourage you to typeset your problem sets and recommend LaTeX, doing so will earn up to 24 bonus points. Typesetting is a useful skill, and will also help us grade more efficiently. 
    • Problem sets must be turned at the beginning of class (12:30pm on Wednesdays) in either hard copy or digitally through a shared dropbox folder. Code, submitted through dropbox and included in the problem set write up, will usually be requested. 
    Your grade will be determined by 7 weekly problem sets, and 2 exams.

    The points will be as follows: 
    7 problem sets worth 50 pts each = 350 pts
    2 exams (1 midterm + 1 final) worth 100 pts each = 200 pts
    Max points: 550 pts

    Bonus: 10 pts for typesetting first problem set
    + 2pts for typesetting subsequent 7 problem sets 
    = 24 total bonus points from typesetting possible (not to exceed 600 max points)

    Group work and Academic Honesty policy: 
    You are encouraged to discuss and work in groups to solve problem sets. 

    You must write up your own solution and your own code. Copy, pasting, and editing will be considered plagiarism. Do not be a cheater, it does not help you learn the material and I will have you do something harder to make up the grade, give you a zero, and/or report you for academic misconduct depending on the situation.

    Please read the UW policy here. By staying registered in the class you indicate your acceptance of all its terms. We do not accept late homework or absence without official reasons (medical, etc.) approved by a student dean. If you miss class, please coordinate with colleagues to find out what you missed (do not email the professor to help you catch up).