🎉 Caltech SURF 2025

Overview

The Summer Undergraduate Research Fellowships (SURF) program is one of the “crown jewels” of Caltech. Since 1979, SURF students have had the opportunity to conduct research under the guidance of experienced mentors working at the frontier of their fields.

In 2025, I will serve as a mentor in the SURF program, co-mentoring alongside CMS postdoctoral researcher George Stepaniants. Together, we will guide a team of 2 to 3 undergraduate students.

Research Focus

The research project in the SURF program will address accessible yet meaningful problems in data assimilation (DA), particularly in combining DA with machine learning methods. The project is designed to be manageable, and I will provide detailed guidance, including theoretical background in both DA and ML, as well as relevant code resources.

My hope is to collaborate closely with the students to produce a research paper by the end of the project. However, if our results are not substantial enough for a publication, that is completely fine. My primary goal is for students to gain a solid understanding and valuable skills in DA and ML. Ultimately, if you walk away from this project having learned a great deal, I will consider it a success.

Expectations for Students

My primary expectation is that you bring genuine enthusiasm to this project. Please make sure that the general research direction outlined above aligns with your interests and that you’re willing to dedicate time and effort to it over the summer of 2025.

In terms of skills, I hope you have a foundational understanding of probability, stochastic processes, (ordinary) differential equations, and linear algebra—ideally having taken coursework in these areas. Coding proficiency is also important; Python experience is essential, and familiarity with or experience in PyTorch is a plus, though not required—I’m happy to provide guidance on this if needed.

Passion and curiosity are the key qualities I’m looking for. If you’re excited about this topic and ready to engage deeply, I believe you’ll find this project both rewarding and educational.

References

For more information on the SURF program, please refer to the official Caltech SURF page. This page provides additional links to details about the program, including Eligibility, Application Information, Announcements of Opportunity, and FAQs.

For a comprehensive understanding of data assimilation, I highly recommend the following resource:

• Inverse Problems and Data Assimilation: A Machine Learning Approach: available on arXiv.

Acknowledgments

I would like to extend my heartfelt thanks to my supervisor, Professor Andrew Stuart, for his invaluable support in my role as a SURF mentor. His guidance and encouragement have been instrumental in enabling me to participate in this program and share this opportunity with undergraduate students.

License

Copyright 2024-present Bohan Chen.

Released under the MIT license.