Grading and policies
Table of contents
Course grading scheme
The course grading scheme is designed to encourage students to keep up with the course content as it happens, and to join lectures in person if they’re able.
- Homework (40%): four homework assignments on problem formulation, algorithm analysis, and algorithm design
- Project (45%): solve an optimization problem
- Presentation (5%): present in-class on recent optimization research
- Participation (10%): attend lecture, ask questions, engage in discussion, and submit course surveys and in-class quizzes and writing assignments
More details about each assignment follow.
Homework
Homework supports our course goals of learning to model optimization problems (hw1), to make mathematical predictions about the performance of optimization algorithms (hw2 and hw3), and learning to tweak optimization algorithms to adapt them to a specific problem (hw4). Homework will be submitted on Gradescope.
Slip days. We recognize that students are balancing many priorities, and so we make accommodations to allow for late homework. Students have 5 slip days that may be used through the quarter with no grade penalty. At most 3 slip days may be used for a single assignment. For example, you could submit one homework three days late and a second two days late. To calculate slip days, we round up: a homework submitted 24 hours and one minute after it is due will use two slip days. Homework submitted after all your slip days are used will receive a score of 0.
Regrades. You can ask for your homework to be regraded up until two days after grades have been posted. Regrades can increase or decrease your grade. (We will regrade the whole assignment, not just a single question.)
Weights. Although some assignments are more difficult than others, we weight all assignments equally when computing your overall homework score.
Collaboration. The goal of the homework is to help you practice the skills that you’ll use later in this class and - we hope! - later in life. Homework carries weight for your grade to encourage you to spend time on it and think deeply about it. Our collaboration policy is geared to make sure you can get the help you need, and so that by the time you turn in your work you understand what it’s about, how it works, and why it’s important.
Students are allowed - and even encouraged - to collaborate on homework. However, each student must submit their own homework. We ask that you
- Give credit to the people who have helped you: please write on your homework the names of the people you worked with.
- Give credit to the other resources that have helped you: please write on your homework the textbooks, notes, web pages, or large language models you found useful.
- Write up your homework by yourself. That is, all of the text that you submit should be typed by you. Tab completion is fine (eg, using Github Copilot).
Partial credit. If you’re not able to answer a homework question, but you show us the work that you performed to think about the question and to try to understand it, you will receive partial credit.
Sharing solutions. Under no circumstance should you seek out or look at solutions to assignments given in previous years, or share or post solutions (yours or ours) to a public website.
Projects
The course project supports our course goal of developing confidence as an optimizer by designing a solution to an optimization problem. You will chooce between three types of projects for this class:
- Optimization methodology
- Detailed application to real-world problem
- Develop LLM tools for optimization modeling
You will work on the project in a team of 1-3 people. (A 1-person team is only acceptable if your project is aligned with ongoing research for your PhD.) You project should be chosen in consultation with the course staff. For PhD students, we suggest choosing a project that to align your work on the class with your longer term research objectives. Come chat with any of the course staff during our office hours to clear your project with us.
As part of the project, you will submit a project proposal, midterm report, and final report, present your findings live during the last week of class, and get (and give) feedback from your peers in the class. The project expectations details project deliverables and lists project ideas.
You will submit your project reports as a pdf, which can be prepared in LaTeX or printed from a Jupyter notebook. We will share projects with other students enrolled in the class for peer-grading, to support our course goals of assessing optimization literature and developing confidence as an optimizer. If you need to keep your project private, please speak with the course staff.
Project final reports will be graded on the following items: (For a methodological or theoretical project, the “problem” referenced below might be that previous literature does not handle problems with certain features.)
- Does the project pose an interesting problem?
- Does the project explore the impact of at least three of the major problem features discussed in class (size, sparsity, discrete variables, convexity, ill-conditioning, access patterns to problem data, convergence tolerance, speed, generalizability)? Which?
- Does the project draw conclusions? Do you believe the conclusions?
- Do the graphics and tables presented support the conclusions of the project?
- Is the writing clear and engaging?
The total project grade will holistically consider the quality and timeliness of the feedback you provided to other students on their projects, as well as your project report and presentation.
Presentation
Students will design short presentations on recent literature in optimization once during the quarter. Sign up for a date and paper to present.
Participation
Students are expected to attend class and actively participate in discussion. We will variously use polling questions and in-class surveys, quizzes, presentations, and writing assignments to quantify participation. Students should expect to receive full participation credit if they miss no more than three days of class.
Exceptions
Beyond the slip days and drops outlined above, extensions on assignments will be granted only with an academic accommodation letter from the Office of Accessible Education (generally for medical reasons). or in other such exceptional circumstances. Requests due to job interviews, other classes and assignments, and poor planning will not be considered. We suggest you save your slip days to insure against catastrophe.