12/13/2022 Julia Park
Written by Julia Park
Research Experiences for Undergraduates (REUs) have become a popular and valuable way for undergraduates from minority-serving institutions to see first-hand the many research opportunities offered at the University of Illinois. Several of MechSE’s own faculty have participated as mentors in REU programs over the years.
This summer, MechSE’s Seid Koric and Chenhui Shao served as mentors and advised the research of four undergraduate students on the topic of “An integrated sensing, machine learning, and high-performance computing framework for real-time decision-making in smart manufacturing.” Theirs was one of 10 projects under the REU-FoDOMMat program (The Future of Discovery: Training Students to Build and Apply Open Source Machine Learning Models and Tools), held at the National Center for Supercomputing Applications (NCSA) and sponsored by the National Science Foundation. Professor Volodymyr Kindratenko, Director of NCSA’s Center for Artificial Intelligence Innovation, was PI of the REU-FoDOMMaT program.
In November, the REU-FoDOMMat program won the “Editors’ Choice: Workforce Diversity & Inclusion Leadership Award” from HPCWire at the Supercomputing Conference 2022. 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. HPCwire is the No. 1 news and information resource covering the fastest computers in the world.
Students in Shao and Koric’s group learned about the global data revolution taking place thanks to recent developments in sensing, communication, and computing technologies and infrastructure. This has resulted in an unprecedented opportunity for the manufacturing industry to adopt a new generation of digitalization and intelligence. As part of the project, the undergraduates developed an integrated sensing, machine learning, and high-performance computing (HPC) framework for next-generation manufacturing process control, and learned about new deep learning algorithms such as convolutional neural network (CNN) and residual neural network (ResNet) for decision-making tasks.
“We were very pleased with the work done by the students in Professor Shao and Professor Koric’s group. They were able to learn a rather complex tool, convolutional neural networks, and successfully apply it to a domain-specific problem. Mentorship by faculty is the key in enabling REU students to succeed in such programs,” said Kindratenko.
Koric is a Research Associate Professor in MechSE and the Technical Associate Director at NCSA. His work focuses on artificial intelligence, large-scale multiphysics modeling, materials processing, advanced manufacturing, high-performance computing, and biomechanics.
Shao is an Associate Professor who conducts research and innovation activities to enhance the automation and intelligence of manufacturing, developing methodologies from a wide range of disciplines, including machine learning, statistics, and automatic control.