Outcomes from National Science Foundation Funding
NRT Stats
53 trainees
Multi-faceted mentorship, peer relationship building, and interdisciplinarity contributed to better retention of graduate students toward degree completion.
24 Peer-Reviewed Publications
Publications from this project covered topics of foundational chemistry, machine learning for catalysis, novel technology development, and improvements to energy storage.
1 Expanded Portfolio
The CEBC’s and KU's expanded research portfolio in machine learning and artificial intelligence for chemical sciences demonstrates our strength in cutting edge research and readiness to prepare the next generation of innovators.
Specific Outcomes
- Trainees determined which machine learning strategies were most appropriate for mining research literature and used these techniques to identify novel experimental parameters for future experiments.
- Trainees demonstrated proof-of-concept for using machine learning to identify novel experimental parameters for carbon dioxide and nitrogen reduction. They also found that shallow learning models are effective for extracting information from published literature.
- Publications from this project covered topics of foundational chemistry, machine learning for catalysis, novel technology development, and improvements to energy storage.
- With better training in how to harness the plethora of data, trainees were able to demonstrate a method to tease out transformative solutions to challenges in the chemical and energy industries.
- Trainees received formal and informal training on highly desirable non-technical skills for the workforce including communication skills, project management, and multidisciplinary collaboration.
- Multi-faceted mentorship, peer relationship building, and interdisciplinarity contributed to better retention of graduate students toward degree completion.
New Directions
- This project laid the groundwork for developing novel data mining and extraction methodologies, which in turn will accelerate catalytic insights and innovations with potentially far-reaching advances in challenging chemistries such as using water and carbon dioxide as energy sources.
- Decades of research seeking to activate the intramolecular bonds in these examples has led to thousands of publications, yet game-changing catalysts remained elusive. The NRT identified machine learning strategies to extract these experimental parameters for future experiments.
- The research advanced the use of machine learning in chemical sciences and demonstrated practical uses for implementing artificial intelligence to identify and optimize catalysis experiments.