ChE Ph.D. Student Siqi Chen Publishes and Presents “Spotlight” Paper at AI Conference
Content
Siqi Chen, a Ph.D. student in the UMass Amherst Chemical Engineering (ChE) Department, has published a peer-reviewed, first-authored paper accepted by the AI for Accelerated Materials Design Workshop at the 38th Annual Conference on Neural Information Processing Systems (AI4Mat Workshop), held on December 14th in Vancouver, British Columbia. Impressively, Chen’s paper was also selected by the AI4Mat Workshop as a “Spotlight” talk, which she presented at the conference in Vancouver. You can find the paper here.
Chen’s Spotlight paper is titled “Integrating Graph Neural Networks and Many-Body Expansion Theory for Potential Energy Surfaces.” The paper introduces a pioneering new framework that takes “a significant step towards computationally aided design of large functional materials,” as her paper explains.
Chen does her research in the Zhou Lin Group in Quantum Chemistry. Zhou Lin is an assistant professor in the UMass Amherst Chemistry Department and an adjunct in the ChE department. Lin’s lab focuses on the development and application of quantum-chemical approaches for complex systems.
Chen’s paper describes her research as a response to a critical problem. As she says, “Rational design of next-generation functional materials relied on quantitative predictions of their electronic structures beyond single building blocks.” However, “First-principles quantum-mechanical (QM) modeling became infeasible as the size of a material grew beyond hundreds of atoms.”
Chen’s research promises to resolve that sizable QM issue. As she says, “In this study, we developed a new computational tool integrating fragment-based graph neural networks (FBGNNs) into the fragment-based, many-body expansion (MBE) theory…and demonstrated its capacity to reproduce full-dimensional, potential-energy surfaces (FD-PES) for hierarchic chemical systems with manageable accuracy, complexity, and interpretability.”
Chen’s paper goes on to explain that “In particular, we divided the entire system into basic building blocks (fragments), evaluated their single-fragment energies using a first-principles QM model, and attacked many fragment interactions using the structure-property relationships trained by FBGNNs.”
Chen’s paper concludes that “Our development of [this integrated FBGNN-MBE computational tool] demonstrated the potential of a new framework, integrating deep-learning models into fragment-based QM methods, and marked a significant step towards computationally aided design of large functional materials.”
Before coming to the UMass Amherst ChE department, Chen earned her B.S. in Energy Chemical Engineering from Beijing University of Chemical Technology in China, her B.S. in Chemical Engineering from Rutgers University in New Jersey, and her M.S. in Chemical Engineering from The Pennsylvania State University.
The AI4Mat Workshop provided an inclusive and collaborative platform in which AI researchers and material scientists converged to tackle the cutting-edge challenges in AI-driven materials discovery and development. As the website explains, “Our goal is to foster a vibrant exchange of ideas, breaking down barriers between disciplines and encouraging insightful discussions among experts from diverse disciplines and curious newcomers to the field.” (January 2025)