Math 294: Optimization

Instructor	Tanya Leise
Email	tleise at amherst dot edu
Phone	542-5411
Office	SMudd 503
Office hours	TTh 10am-noon MWF 1-1:50pm Other times by appointment or simply try stopping by my office
Text	Optimization Models (Available at Amherst Books and on reserve in the science library)

Course goals:

• Become proficient in constrained optimization and associated linear algebra
• Understand both theory and applications
• Explore examples using computational software

Course Topics:

• Linear algebra used in convex optimization
• Convexity
• Linear and quadratic programs
• Second-order cone programs
• Semidefinite programs
• Applications

Attendance: Please be in class and be there on time. Cooperative learning is more effective and more fun than struggling through material on your own. If you do miss a lecture, it is your responsibility to obtain the material that you missed and to get your assignments handed in to me.

Questions: If you have a question during lecture, please raise your hand and ask it right away. Chances are that other students are wondering the same thing. If a question arises later, feel free to visit my office and we'll work through sample problems until you are comfortable with the mathematics. Always feel free to ask me to slow down as well.

Numerical software: We'll be doing lots of numerical explorations and learning to implement algorithms. If you have a laptop, please bring it to class regularly; if you don't have a suitable device, please let me know and we'll try to arrange one for you to borrow from the college. We will use Mathematica in class to explore examples. The college has a license that should allow you to install Mathematica on your own computer, and it's available on most college computers.

Grading: Your course grade will be based on three take-home exams (50% total), homework (30%), and a final project with a report and presentation (20%).

Intellectual Responsibility

• Exams. Your work must be entirely your own, so please follow the guidelines of the honor code.
• Homework. You may study with other students following these guidelines:

1. If you worked with or received help from any source other than me, you should put a note on the front of your homework saying, "I worked with <names>." Make sure your name stands out as the author of your homework.
2. Working together does not mean that one of you does the first half of the homework set and the other does the second. Everyone should work on every problem.
3. Each student must hand in his or her own problem set. You may not hand in a single packet as the work of multiple people. You must do your own write-up of each problem.
4. Do not copy someone else's solution—you will not learn anything and it is plagiarism. You may discuss problems with others, but then you must be able to work out the solution on your own again and write it down yourself.
5. Sending or receiving a copy of a file that contains student work (paper, lab, homework assignment, etc) violates the honor code and will be treated as plagiarism. You may talk with others about strategies for solving a problem, but please do not share files.
6. Be cautious in searching the web or other resouces for homework help. Copying solutions is plagiarism, whatever the source.
7. If you are unsure what agrees or does not agree with the precepts of intellectual responsibility in this course, feel free to talk to me about it.

Homework Guidelines

• All problem sets are due at THE START OF CLASS. Late homework will receive half credit for homework handed in after start of class but by the start of the next class meeting (and will not accepted after that, unless you contacted me in advance to obtain an extension).
• If you are unable to attend class due to illness or an emergency, let me know as soon as you can and we will work out an appropriate schedule for assignments.
• Your name should be written on all sheets handed in.
• Problem solutions must be written out in the order they were assigned.
• Multiple pages must be STAPLED.
• Homework should be neat. No dog ears. No messy edges from notebook paper.
• Where appropriate, please box or highlight final answers. In general, try to make your answers readable and easy to find. Always keep the grader happy!
• As mentioned elsewhere, no copying! Always keep the honor code in mind to maintain an ethical environment in which everyone truly learns the material.

Course Resources:

Don't struggle alone! You have many options for getting help with this course.

• Me. Feel free to come to my office hours, make an appointment by email or phone, or simply try stopping by my office—you are welcome whenever my door is open. If you have some anxiety about taking math exams, please come see me and we can work together on building your math confidence.

• Homework. Mathematics is learned ACTIVELY, not passively. You can't absorb math through listening or reading, even if you think you understand it all.
• Textbook. I won't go over everything that is contained in the text, and I try to avoid doing the same examples. Hence your textbook in an important independent source of information and you should read it!
• Lecture notes. Reviewing the notes you take in lecture will give you a chance to see the material again after you have had some time to assimilate it.
• Your classmates. Discussing math with others can help you think through the concepts. Explaining an idea you already understand will deepen your comprehension, and for the concepts that you don't understand well, the explanation of a peer may be more helpful than mine or the textbook's.
• Library resources.
• The textbook is on reserve in the Keefe Science Library.
• The library has other books on numerical analysis and scientific computation that may be useful resources to supplement the coverage in our text.
• Mathematica can be installed on your own computer (https://www.amherst.edu/offices/it/knowledge_base/software/college-install) or accessed through a Remote Desktop Connection (https://www.amherst.edu/offices/it/knowledge_base/network-wifi/Remote_Desktop_Connection). The college has a limited number of Mathematica licenses, so be sure to quit Mathematica when you are finished; otherwise, you may prevent another student from being able to use it.

To the Math 294 Homework Schedule