CMS 273, Winter 2021
Frontiers in Computing and Mathematical Sciences
- Organizers: Richard Murray (firstname.lastname@example.org), E. Schmidt, S. Feldman
- Organizational meeting: TBD, but most likely 4 or 6 Jan (Mon, Wed), 12 pm via Zoom
The purpose of this course is to explore applications of tools from Computing and Mathematical Sciences to new problem domains. The course is organized around small teams consisting of CMS and non-CMS students who will work on projects of mutual interest in some faculty member's research area. Our main goals are for the participating CMS and science/engineering faculty to become more familiar with each other's work and expertise, and to get our graduate students interacting with one another.
The output of the course will be a short paper of the sort that could be sent to a conference. The paper should consist of a short description of the problem under study and the relevant CDS tools, followed by a preliminary set of results and a description of next steps to be pursued.
|1||4 or 6 Jan||Organizational meeting, 12 pm via Zoom|
|8 or 11 Jan||First team meeting, 12 pm via Zoom|
|2-4||Work in teams; define problem to be studied + approaches|
|5||1-5 Feb||Midterm presentations, time TBD via Zoom|
|6-8||Work in teams|
|9||8-12 Mar||Final presentations, time TBD via Zoom|
|11||17 Mar||Final reports due (by 5 pm)|
The projects for this years class are listed below. To access the project descriptions, you must be on the Caltech network (on campus or via VPN).
- Michael Dickinson (BBE): Modeling and control of the effects of bilateral asymmetries in insect flight (coming soon)
- Frederick Eberhardt (HSS): Causal discovery of the emotional state of people during the COVID pandemic
- Bethany Ehlmann (GPS): Extracting maximum compositional information from hyperspectral Images for Earth and planetary exploration
- Christian Frankenberg (GPS): Detecting and quantifying methane sources with AI
- Matt Thomson (BBE): Real-time machine learning with self-organized neural networks: a bio-inspired take on grand challenges in AI
- Magda Zernicka-Goetz (BBE): Migration of the anterior signaling centre to specify location of the head / Building a developmental atlas of mammalian gastrulation - a dynamic molecular colouring book
- A Survey of Deep Learning for Scientific Discovery, M. Raghu and E. Schmidt
Units and Grading
CMS 273 is a 9 unit course, offered either graded or pass/fail. Each team is expected to complete the following:
- Project presentation: each team will make short (15-20 min) presentations in the middle and toward the end of the term, describing the focus of their project. Comments on these presentations will be provided to the team for incorporation in the final report.
- Final report: each team will prepare a paper describing their work during the term. This should build on the midterm report by including some preliminary results and/or case studies.
In order to complete the work for the term, each team should plan on meeting at least once per week. The first team meeting will be on either Friday, 8 Jan, or Monday, 11 Jan, at 12 pm via Zoom (at which time a regular meeting time can be established by the team).
Frequenty Asked Questions
Q: Can I propose my own project to work on?
A: Not at the present time. Because we need to find projects that have the right CMS content and that have a faculty sponsor and subject matter experts available, we line up the projects prior to the start of class. If you have an idea for a project that we might offer next year, send e-mail to Richard and he'll put it on his list.
Q: How will the project work during COVID?
A: Most activities will happen remotely, but project teams can use Caltech's outdoor meeting spaces to meet in person if everyone is n the Pasadena area (and subject to the limitations around the use of those areas).
Q: Can I still participate if I can't make it to the first class?
Yes! Just make sure to send e-mail to Richard so that you get the link for the preferences survey and that you submit it on time.