Can Li, Co-Director of the Purdue Center for Operations and Optimization in Process Systems (COOPS) and Assistant Professor in the Davidson School of Chemical Engineering, has been awarded a prestigious NSF CAREER Award. This funding will support his research in process systems engineering and machine learning.

 

The project, Novel Neural Network Architectures Inspired by Optimization Algorithms, will receive $500,000 in funding over the next five years. This award supports Li's innovative research on integrating optimization principles into machine learning models to rigorously enforce physical and logical constraints, thereby enhancing their reliability and interpretability.  

 

Machine learning, particularly deep neural networks, has revolutionized fields like chemical engineering and process systems engineering (PSE) but is often hindered by its "black-box" nature, making outputs difficult to interpret and noncompliant with physical laws. To address this, Li's project develops optimization-inspired neural networks (OINNs), a novel architecture that integrates optimization theory to enforce strict physical and logical constraints. This approach enhances model reliability, accuracy, and applicability in areas such as process design, control systems, and chemical structure prediction.

 

The educational initiatives associated with this project include redesigning an existing data science course at Purdue to integrate AI and optimization principles tailored to chemical engineering. Li will also recruit undergraduate researchers through the Research Experiences for Undergraduates (REU) program and engage K-12 educators and students through innovative outreach efforts, such as a video game illustrating engineering principles and STEM-focused teacher workshops.  

 

The NSF Faculty Early Career Development (CAREER) Program is one of the National Science Foundation's most prestigious initiatives, recognizing early-career faculty who demonstrate exceptional potential to excel as academic role models in research and education. The program supports projects that establish a strong foundation for integrating innovative research with impactful educational activities, fostering a lifetime of leadership and advancement in their field. To learn more about the NSF CAREER program, visit the NSF website.

 

Li joined Purdue Chemical Engineering in 2022. The Li Research Group focuses on the development of theory, algorithms, models, and software for large-scale optimization and machine learning with applications in energy, chemical, and biological systems. In 2023, Li was awarded the Scientific Prize and Research Contract from Air Liquide for his innovative work in optimization and machine learning. In 2024, Li was awarded the Amazon Research Award and the ACS PRF Doctoral New Investigator Award.

Learn more about Professor Can Li at https://engineering.purdue.edu/ChE/people/ptProfile?resource_id=262693