Grier Jones, fifth year chemistry PhD student, won a poster competition at the spring meeting of the American Chemical Society (ACS). His poster, titled “Exploring the topology of electronic correlation with graph neural networks” earned the NVIDIA GPU Award for Best GPU Poster. The award targets excellent computational chemistry research using a graphical processing unit (GPU).
GPUs are most often associated with the high-quality images seen on gaming computers. However, the highly parallelized architecture of GPUs offers an acceleration platform that can outperform central processing units (CPUs) when processing large amounts of data in parallel. This has implications for scientific computing and machine learning applications, which have traditionally used CPUs.
Jones has developed a novel computational model that incorporates GPUs with graph neural networks (GNNs) and topological data analysis (TDA) to explore the topology of electron correlation. Jones notes this project is unique in that it allows him and fellow researchers to look at electron correlation in the context of machine learning from a new perspective.
Jones joined the UT Department of Chemistry as a graduate student in 2018. He began working with Associate Professor Konstantinos Vogiatzis, whose lab and research group supported Jones and helped develop the work featured in his award-winning poster.
“It was amazing to win this award because there are many successful scientists in our field, a few of which I know personally, that have won this award,” Jones said. He went on to express his gratitude for the Graduate Student Senate Travel Award and Vogiatzis’ NSF-CAREER award, which made it possible for him to participate in the ACS Spring 2023 meeting and the poster competition.
Shortly after winning the poster award, Jones was also named a Gleb Mamantov Graduate Chemistry Scholar by the Department of Chemistry. The poster award, which provided a professional workstation-level NVIDIA GPU, and the Mamantov prize allowed Jones to build an exceptional PC.
“I am very grateful for both awards because building my own PC was a dream come true that I did not think I was going to be able to do until after graduate school,” said Jones. “I have it set up for both work-related computational tasks, which I run daily, and Windows, which lets me edit documents with Microsoft Office and do some gaming.”
Jones’ research was developed with GPUs provided by the Infrastructure for Scientific Applications and Advanced Computing (ISAAC) cluster at UT. Jones described the opportunities and support available at the university as a great environment that has contributed to his intellectual growth and academic exploration.
Prior to joining the chemistry department at UT, Jones earned his undergraduate degree at the College of Charleston, where he became very passionate about computational science as it applies to machine learning and chemistry. Pursuing the Interdisciplinary Graduate Minor in Computational Science and working with Vogiatzis has allowed him to continue exploring these research areas.
“Professor Vogiatzis has really pushed me to new heights, while allowing me to integrate my passions into our projects. I would say this is exactly what any student would want from their graduate school experience; freedom, intellectual satisfaction, and recognition of their achievements,” said Jones.
Since joining the university, Jones has co-authored three publications in journals such as the Journal of Physical Chemistry Letters and Inorganic Chemistry Frontiers, and has contributed to the book, Molecular Representations for Machine Learning.