ML Batched Superhydrides
Template-guided machine learning for discovering stable complex metal superhydrides and prioritizing promising high-pressure superconducting candidates.
M.S. Chemistry candidate at CSU Northridge · Ph.D. applicant in theoretical & computational materials science
I am a Master of Science candidate in Chemistry at California State University, Northridge, where I work in the Miao Lab on theoretical and computational materials chemistry. My research centers on 2D materials, chemical behavior of materials under pressure, and machine-learning acceleration of electronic-structure workflows for novel materials discovery.
I design automated, high-throughput pipelines for quantum chemistry calculations, maintain our lab’s GPU-enabled HPC resources, and translate results into reports for collaborators, DoD sponsors, and NSF programs.
I am excited to lead this semester’s first Chem/Biochem Journal Club meeting on February 20. We will discuss Metallic Oxides and the Overlooked Role of Bandwidth, a very interesting Chemistry of Materials paper that connects solid-state chemistry with core electronic-structure concepts. If you are curious and want to attend, please reach out by email; beginners are absolutely welcome.
Thank you to the UCSB team for such a fascinating piece of research.
I received a DCOMP travel award to attend the Global Physics Summit 2026 in Denver, Colorado. I will be presenting my electron localization function (ELF) work that I first shared at NeurIPS AI4Mat in December.
In late January, I attended the Gordon Research Conference on Multifunctional Materials and Structures in Ventura, CA. I presented two posters: one on machine-learning-guided discovery of complex metal superhydrides and one on symmetry-aware deep-learning prediction of electron localization functions. It was an excellent chance to exchange ideas across disciplines and build new research connections.
Our lab's two most recent publications were featured on the CSUN Department of Science and Mathematics news site: one highlighting our machine-learning guided materials discovery work (article link) and another covering pressure-induced redox reversal of iron in deep-Earth conditions (article link).
I am proud to announce that I received a travel grant to attend NeurIPS AI4Mat 2025 and will be presenting two posters there. One shares our JACS study on batch discovery of complex metal superhydrides and how template-guided ML accelerates the search for high-pressure materials (PDF). The other covers an in-progress symmetry-aware 3D-UNet that predicts electron localization functions from superposed atomic density grids with promising early accuracy (PDF).
Our group's work on pressure-driven redox reversal of iron and its role in element distribution deep in Earth was featured by CSUN Newsroom. The piece highlights how computational evidence for iron's redox reversal points to new bonding pathways for p-block elements under extreme pressure and can reshape models of core formation and evolution. Read the story.
A curated set of recent posters spanning machine-learning-guided materials discovery, high-pressure chemistry, and symmetry-aware electronic-structure modeling.
Template-guided machine learning for discovering stable complex metal superhydrides and prioritizing promising high-pressure superconducting candidates.
DFT analysis of pressure-driven metal-insulator transitions in molecular hydrogen and how evolving electronic delocalization relates to superconductivity pathways.
A periodic, symmetry-aware deep-learning model that predicts electron localization functions from superposed atomic densities for fast electronic-structure screening.
Vacancy-controlled electride behavior in 2D transition-metal dichalcogenides and the impact of localized interstitial electrons on catalytic hydrogen-evolution trends.
First-principles structure-search results showing pressure-induced breakdown of polyoxygen motifs in Cs-O compounds and activation of Cs core-level electrons in high-pressure bonding.
X. Wang, X. Feng, J. Li, Y. Lv, A. Ellis, S. Scott, A. Pandit, D. Khodagholian, R. J. Hemley, M. G. Jackson, F. J. Spera, S. A. T. Redfern, M. Miao. Proceedings of the National Academy of Sciences 122, e2414911122 (2025). DOI 10.1073/pnas.2414911122.
Y. Sun, A. Ellis, X. Chen, M. Miao. Journal of the American Chemical Society 147(44), 40407–40419 (2025). DOI 10.1021/jacs.5c11731.
W. Zhao, Austin Ellis, D. Duan, H. Wang, Q. Jiang, M. Du, T. Cui, M. Miao. Advanced Functional Materials 35 (2025) 2415910. DOI 10.1002/adfm.202415910.
Y. Sun, Austin Ellis, S. Diaz, W. Li, M. Miao. The Journal of Physical Chemistry Letters 15 (23), 6174–6182 (2024). DOI 10.1021/acs.jpclett.4c01263.