AIDMM-NRT

Complex fluids, biological systems, and multi-component materials are examples of challenging multi-scale problems that are of critical importance for society at large, playing an important role in modern life from computer chips to new cancer therapies, to lightweight materials in cars and aircraft. Despite an enormous multi-decade national investment in massively parallel computational resources and scalable numerical methods, these problems remain some of the most difficult and challenging to solve over the length and time scales of interest. To meet this challenge across a range of important applications, the participants in this NRT come together to advance the fundamental understanding of critical problems by integrating high-fidelity computed and experimental data to create data-informed predictive surrogate models based on modern AI inspired techniques coupled with traditional simulation tools.

Learn more about the program!

---

What is the AIDMM-NRT Program?

The AIDMM-NRT program is led by Michigan State University’s Department of Computational Mathematics, Science and Engineering in collaboration with faculty members from the Department of Mathematics, the Department of Statistics and Probability, the Department of Biochemistry and Molecular Biology, the Department of Physics and Astronomy, and the Institute for Cyber-Enabled Research. It establishes a world-class research and training program across these units to Harness the data revolution to enable predictive multi-scale modeling across STEM. The program will address two of NSF’s 10 Big Ideas namely HDR and AI. The dual research and training thrusts will:

1. Develop predictive multi-scale models by novel structure preserving machine learning methods in three core areas: 
   1. complex fluids
   2. polymeric materials
   3. biophysics
2. Create an innovative, flexible, and dynamic environment for interdisciplinary graduate education that emphasizes scientific writing and communication, critical thinking, and career development; 
3. Develop a cross-department curriculum including core courses in structure preserving ML, HPC, and scientific communication, as well as targeted short courses in cutting-edge topics. 

Our trainee cohort will include graduate students from the member departments, but the components of the program will be open to students from other science and engineering disciplines. The program features several innovative aspects. Graduate assistantships funded by this training program will be allocated to MSU PhD students whose research focuses on any aspect of SP-ML, including, but not limited to, the development of: new algorithms for existing SP-ML; theory to provide insight into key challenges such as those described in the research section below in SP-ML; and in the applications we consider here. 

Apply here!

Support will be prioritized for those who come from groups underrepresented in STEM. All aspects of the training program will be open to any PhD students at MSU, to the greatest extent possible. A salient feature of this training program will be substantial engagement with partners in industry and in the national laboratories, who will participate in mentoring students in research and professional skills as well as place students into internship 

---

Program Features

The AIDM-MSU-NRT has several key features that together will create an innovative, flexible, and dynamic environment for interdisciplinary graduate education. Key features of the program are: 

• core coursework leading to a graduate certificate, 

• core course on scientific communication, 

• targeted short courses in cutting-edge topics, 

• an internship at a partner site, 

• short-term visits from partners, 

• a professional development program, 

• inclusion of partners in PhD students’ guidance committees.

---

Core Program Elements

• MSU Graduate Certificate in SP-ML - the Foundational NRT Courses 

  • CMSE 820 Mathematical Foundation of Data Science

  • High Performance Computed High Fidelity Data 

  • Structure Preserving Physics-enforcing Surrogate Models 

• Targeted Short Courses on Cutting-edge Topics

• Internship Program 

• Partner Visitor Program 

• Professional Development 

• Leveraging Existing MSU Resources 

• Mentoring of Junior Students 

• Thesis Topic and Guidance Committee 

• Partner Institutions 

---

Timeline

Trainees admitted to the NRT cohort will have a diverse background, including Computational Mathematics, Science and Engineering (CMSE), Mathematics, Statistics and Probability, Biochemistry and Molecular Biology and Physics and Astronomy. Trainees need to fulfill the PhD requirements of their home departments. As such, students in the first year will only be expected to attend the scientific communication seminar and the regular AIDM-MSU-NRT seminar. Ambitious students may take Course 1 in their first year as well.

Students will be expected to complete the NRT curriculum by the end of their second year. The sequence will run every year, with Course 1 in the fall and Courses 2/3 in the spring. Students will be required to complete the internship no later than the start of their fifth year, but will be encouraged to complete it between their second and third years. This timeline fits the requirements of the participating departments.