The Department of Mathematics and Computer Science at Emory University is looking for two undergraduates who are strongly motivated to develop new mathematical and computational methods for deep learning. The students will be hosted on Emory’s main campus in Atlanta, GA, for 6-8 weeks during the summer (start and end dates are flexible). We particularly encourage women, students from underrepresented groups or institutions, and first-generation college students to apply.
Participants will work with Dr. Lars Ruthotto and will be introduced to new approaches to designing deep neural networks by modeling them as dynamical systems. This interpretation leads to intriguing computational and analytical questions that the students will address under the PI’s supervision.
Applicants should bring a strong background in applied mathematics (including numerical linear algebra, differential equations, optimization, …) as well as technical skills (most importantly, programming skills in MATLAB or Julia). Other optional but highly valued skills include parallel computing, GPU computing, partial differential equations. In addition to writing a paper, students are expected to contribute to a new high-quality open source framework for deep learning developed by the PI and collaborators.
The program is supported by the National Science Foundation award DMS 1751636. Supported participants will receive a stipend of 3,500 USD, travel support of up to 1,000 USD, and free accommodation.
More information about the PI can be found here: http://www.mathcs.emory.edu/~lruthot/
More information about Scientific Computing at Emory can be found here: http://www.mathcs.emory.edu/Research/Area/ScientificComputing/
- Curriculum Vitae
- Research Statement
- Transcript (unofficial ok)
- Two Reference Letters (to be submitted by the reference writers at this site)
Application can be submitted at: https://www.mathprograms.org/db/programs/659
- Undergraduate News