Eight undergraduate students completed a successful 10-week summer research experience in machine learning and open-source software and model development at the National Center for Supercomputing Applications.
The students participated in “The Future of Discovery: Training Students to Build and Apply Open-Source Machine Learning Models and Tools” (FoDOMMaT) in which they worked on developing open-source software and models then applied them to solving real-world problems. The internship program – made possible through a $405,000 National Science Foundation (NSF) award – was part of a larger student engagement program hosted by NCSA.
“The program started with a four-day tutorial in machine learning delivered by graduate student Priyam Mazumdar whom we involved for the summer to act as a resident expert,” said Volodymyr Kindratenko, director of the Center for Artificial Intelligence Innovation and principal investigator on the NSF grant that funded this summer program. “The students then went ahead with the projects from their faculty mentors while Priyam continued to provide expert advice with the methods and tools. We were very impressed by the work carried out by our Research Experiences for Undergraduate (REU) students over the summer.”
This cohort of students, a majority of whom are part of historically underrepresented communities in STEM, worked on cutting-edge projects, including the estimation of crop productivity from multi-sensor fused satellite data and designing new materials through machine learning. In addition to the rigorous summer training and research, FoDOMMaT students had opportunities to participate in professional development seminars offered twice a week and socialized with NCSA and students from other REU programs on campus.
The excellence of the FoDOMMaT program and CAII were recognized with an HPCwire Editors’ Choice Award, winning the “Workforce Diversity & Inclusion Leadership Award” at the 2022 Supercomputing Conference in November.
The coveted HPCwire Readers’ and Editors’ Choice Awards are determined through a nomination and voting process with the global HPCwire community as well as selections from the HPCwire editors. The awards are an annual feature of the publication and constitute prestigious recognition from the HPC community.
Diversity and inclusion are central to the FoDOMMaT program’s goals and structure, and we are committed to creating an inclusive culture where all participants feel valued, respected and empowered to perform at their best. We are very pleased to receive this award as a recognition of our efforts.
Volodymyr Kindratenko, CAII Director
The FoDOMMaT cohort collaborated with students from other NCSA programs, including Students Pushing Innovation (SPIN) and the NCSA International Research Programs. Three students from the King Abdullah University of Science and Technology Gifted Student Program worked with FoDOMMaT participant Seenara Khan to develop “An Integrated Sensing, Machine Learning and High-Performance Computing Framework for Real-Time Decision-Making in Smart Manufacturing” project.
SPIN offers paid internships to University of Illinois Urbana-Champaign students – including those without a strong technical background – who have the opportunity to contribute to challenging, interdisciplinary projects with support from members of NCSA’s expert staff who act as mentors.
Students from all three programs were required to develop a research plan at the beginning of the summer and a research report at the conclusion of the program. They gave presentations at joint Lightning Talks scheduled weekly and participated in the 2022 Illinois Summer Research Symposium and the 2022 Summer Joint REU FoDOMMaT/SPIN/NCSA International Research summer poster session on July 29. Students provided valuable contributions that helped to advance their projects.
In fact, five of the FoDOMMaT participants and 12 SPIN interns were recognized for their outstanding contributions with a 2022 Summer NCSA Recognition Letter signed by NCSA Director Bill Gropp. See the list of students and their projects below:
FoDOMMaT
- Morgan Cosillo, “Translating Optical Character Recognition Text to Words with the Reading Time Machine.” Mentor: Jill Naiman
- Natalie Foss, “Comparing Deep Learning and Expert Knowledge for Sequential Pattern Mining.” Mentors: Nigel Bosch and Luc Paquette
- Chancharik Mitra, “Spatial Analysis of Tumor Heterogeneity Using Machine Learning Techniques.” Mentor: Zeynep Madak-Erdogan
- Isabelle Wagenvoord, “Estimation of Crop Productivity from Multi-sensor Fused Satellite Data.” Mentor: Kaiyu Guan
- Jeffrey Zhou, “Design of New Materials Through Machine Learning.” Mentors: Andre Schleife and Michal Ondrejcek
SPIN
- Hao Bai, “Democratizing AI Literacy by Using Physical Programming with Lego Bricks and Keras Framework.” Mentor: JooYoung Seo
- Joshmita Chintala, “Web-interface for Atmospheric Chemistry Simulations.” Mentor: Nicole Riemer
- Anushka Gami, “Quantifying the Effectiveness of Scientific Documentaries Using Natural Language Processing.” Mentor: Jill Naiman
- Yumingxuan Guo, “Music on High Performance Computers.” Mentor: Sever Tipei
- Bo Pang, “Using Machine Learning to Predict Eye Movements in Skilled and Unskilled Readers.” Mentors: Jon A. Willits and Anastasia Stoops.
- Hrishikesh Kalyanaraman, “Continuous Integration Testing for the Einstein Toolkit.” Mentor: Roland Haas
- Joe Lue, “Sonification of Electron Density Data.” Mentor: Andre Schleife
- Yiwen Miao, “AVL Documentary Viewer Counts.” Mentor: Kalina Borkiewicz
- Tonya Muzhylko, “Multiscale Modeling of the Cellular Membrane-Associated Phenomena.” Mentor: Taras Pogorelov
- Linh Pham, “Quantifying the Effectiveness of Scientific Documentaries Using Natural Language Processing.” Mentor: Jill Naiman
- Shane Wu, “Web-interface for Atmospheric Chemistry Simulations.” Mentor: Nicole Riemer
- David Zhu, “Translating Optical Character Recognition Text to Words with the Reading Time Machine: OCR Engine Optimization.” Mentor: Jill Naiman
