Yuan (Cyrus) Chiang

PhD candidate at UC Berkeley and LBNL | Berkeley Fellow | AI/ML + Computational Materials Science

Welcome to my cyber nook! I am Yuan Chiang (江元, pronounced as ‘You-and John’). I also go by my coffee name Cyrus. I am a PhD candidate in Materials Science and Engineering at UC Berkeley and Lawrence Berkeley National Laboratory, under the guidance of Prof. Mark Asta. I leverage density functional thoery calculations and develop machine learning models and frameworks on supercomputers to drive our understanding and innovations in AI for physical sciences. Recently I work on the devlopement and benchmark of foundation machine learning interatomic potential (MLIP) and use them as a probe to understand and design ferroelectric materials and metal-salt interface in Gen IV molten salt fission reactors (MSRs).

My research interests more broadly lie in computational materials physics and chemistry at the atomic and molecular levels, with an emphasis on the theory and application of ab-initio calculations, molecular dynamics, and AI/ML to tackle challenges across energy, materials, pharmaceutics, devices, and computing.


Previously I earned my BS in Civil Engineering and MS in Structural Engineering at National Taiwan University, under the supervision of Prof. Shu-Wei Chang. I was a structural engineering intern at T.Y. LIN in 2018 and was a full-stack intern at Ucarer in 2021.

Yes! I had an unorthodox transition from “civil” to “materials science” (they are at the opposite ends of spectrum in terms of the length scale if you will) after years of working with Prof. Chang in civil engineering department because he is too nice and I was too numb not seeing what I really want to pursue, thanks to the spoon-feeding education system in Taiwan and my self-deception from the uninformed, vague childhood interest in civil structure 🙃. Fortunately, I had Prof. Chang’s unwavering guidance that enlightened me to use my gift and interest for fundamental science and technology. :stars:


News

May 8, 2025 I passed my qualifying exam and become a PhD candidate! :rocket:
Apr 8, 2025 MLIP Arena is accepted as an ICLR AI4Mat Spotlight... Read more
Dec 30, 2023 We are excited to share MACE-MP-0, the foundation ML model for atomistic materials chemistry, on arxiv.
Oct 23, 2023 We release a universal machine lenaring interatomic potential on Huggingface... Read more
Nov 28, 2022 I am giving two talks at MRS Fall 2022 in Boston. Check out the abstracts in the link for details!... Read more

Latest Posts

Selected Publications

2025

2024

  1. ICLR-AI4Mat Poster
    Yuan Chiang, Elvis Hsieh, Chia-Hong Chou, and 1 more author
    arXiv preprint arXiv:2401.17244 2024

2023

  1. 2023-mace-mp.png
    Ilyes Batatia*, Philipp Benner*, Yuan Chiang*, and 8 more authors
    2023

2022

  1. Yuan Chiang, Wei-Han Hui, and Shu-Wei Chang
    Cell Reports Physical Science 2022