Academic background

Sep 2025 – present: Postdoctoral Fellow
The University of Hong Kong Hong Kong Quantum AI Lab Limited
Sep 2021 – Sep 2025: Doctor of Philosophy
The University of Hong Kong
Aug 2017 – Jul 2021: Bachelor of Science
Nankai University

Research interests

AI for Science

Vector conceptual illustration connecting electronic structure, machine learning potentials, and layered battery materials. — Electronic structure Machine learning potentials Materials

Selected publications

PEACE architecture: molecular graph, equivariant interaction layers, parity-resolved Hamiltonian and electronic connection heads, and covariant predictions of energies, forces and nonadiabatic couplings.

PEACE: Covariant learning of nonadiabatic manifolds with parity-resolved Hamiltonians

Rongzhi Gao^* et al.

Preprint* Corresponding author

arXiv: 2610.09576 (2026).

arXiv GitHub code

BibTeX

@misc{gao2026_peace,
  title = {{PEACE: Covariant learning of nonadiabatic manifolds with parity-resolved Hamiltonians}},
  author = {Gao, Rongzhi and Chen, Shuguang and Zhou, Yang and Chen, GuanHua and Hu, Ziyang and Yam, ChiYung},
  year = {2026},
  eprint = {2610.09576},
  archivePrefix = {arXiv},
  primaryClass = {physics.chem-ph},
  doi = {10.48550/arXiv.2610.09576},
  url = {https://arxiv.org/abs/2610.09576}
}

Concept TOC showing a voltage-biased lithium-water interface, equivariant graph encoding, and energy-force prediction.

Voltage-embedded equivariant machine learning potential for open system simulations

Yiming Guan, Rongzhi Gao^*, et al.

Preprint* Corresponding author

arXiv: 2609.04696 (2026).

arXiv

BibTeX

@misc{guan2026_voltage,
  title = {{Voltage-embedded equivariant machine learning potential for open system simulations}},
  author = {Guan, Yiming and Gao, Rongzhi and Yam, ChiYung and Chen, GuanHua and Hu, Ziyang},
  year = {2026},
  eprint = {2609.04696},
  archivePrefix = {arXiv},
  primaryClass = {physics.chem-ph},
  url = {https://arxiv.org/abs/2609.04696}
}

Figure 1 overview of the foundation machine learning potential with polarizable long-range interactions.

A foundation machine learning potential with polarizable long-range interactions for materials modelling

Rongzhi Gao et al.

Nature Communications 16, 10484 (2025).

Paper GitHub code

BibTeX

@article{gao2025_2,
  title = {{A foundation machine learning potential with polarizable long-range interactions for materials modelling}},
  author = {Gao, Rongzhi and Yam, ChiYung and Mao, Jianjun and Chen, Shuguang and Chen, GuanHua and Hu, Ziyang},
  journal = {Nature Communications},
  year = {2025},
  volume = {16},
  number = {1},
  pages = {10484},
  doi = {10.1038/s41467-025-65496-3}
}

The published TOC graphic for delta-learning calibration of lithium-ion battery open-circuit voltages.

Accurate Prediction of Open-Circuit Voltages of Lithium-Ion Batteries via Delta Learning

Wai Yuet Chiu, …, Rongzhi Gao, et al.

Journal of Chemical Theory and Computation 21(10), 5230–5235 (2025).

Paper

BibTeX

@article{chiu2025_delta,
  title = {{Accurate Prediction of Open-Circuit Voltages of Lithium-Ion Batteries via Delta Learning}},
  author = {Chiu, Wai Yuet and Zhang, Chongzhi and Gao, Rongzhi and Hu, Ziyang and Chen, GuanHua},
  journal = {Journal of Chemical Theory and Computation},
  year = {2025},
  volume = {21},
  number = {10},
  pages = {5230--5235},
  doi = {10.1021/acs.jctc.5c00168}
}

Concept TOC showing an all-solid-state lithium battery and electronic band alignment.

Self-Consistent-Charge Density-Functional Tight-Binding Parameters for Modeling an All-Solid-State Lithium Battery

Rongzhi Gao et al.

Journal of Chemical Theory and Computation 19(5), 1381-1387 (2023).

Paper

BibTeX

@article{gao2023_1,
  title = {{Self-Consistent-Charge Density-Functional Tight-Binding Parameters for Modeling an All-Solid-State Lithium Battery}},
  author = {Gao, Rongzhi and Hu, Ziyang and Mao, Jianjun and Chen, Shuguang and Yam, ChiYung and Chen, GuanHua},
  journal = {Journal of Chemical Theory and Computation},
  year = {2023},
  volume = {19},
  number = {5},
  pages = {1381--1387},
  doi = {10.1021/acs.jctc.2c01115}
}