Nuclear Theory · Emergent Symmetries · AI for Science

Phong Q. Dang

I develop symmetry-adapted first-principles methods and interpretable, symmetry-aware AI to discover how collective structure emerges in atomic nuclei—and to make increasingly complex nuclear systems computationally tractable.

Explore my research Download CV
Portrait of Phong Q. Dang
8Publications
5First-author papers
25+Research talks & invited activities
2025Quad Fellow

About

Uncovering the hidden order in atomic nuclei

My work asks how simple organizing principles emerge from the complex interactions of protons and neutrons. I pursue that question through representation theory, symmetry-adapted ab initio modeling, high-performance computing, and machine learning.

I am the main developer of the Symmetry Adapted Model (SAM), have built general U(3) and U(4) coupling algorithms for nuclear theory, and created the Wigner-Informed Neural Network (WINN) to connect first-principles symmetry structure with global nuclear data. My long-term goal is a research program in which physics and AI continually sharpen one another: theory supplies interpretable structure, while AI guides discovery across the nuclear chart.

Education

Higher education

Ph.D. in Physics — Louisiana State University

2023 – present · Baton Rouge, LA

Nuclear structure theory, advised by Dr. Jerry P. Draayer (2023–2026) and Dr. Michela Negro (2026–present). Qualifying exam passed January 2024.

B.Sc. in Physics — University of Debrecen

2020 – 2023 · Debrecen, Hungary

Graduated summa cum laude, advised by Dr. József Cseh. Thesis: Shape Isomers of α-like Nuclei in terms of the Multiconfigurational Dynamical Symmetry.

What I do

A connected program in symmetry, computation & AI

Σ

Symmetry-adapted ab initio models

Developing SAM as a high-performance framework for symmetry-resolved first-principles calculations, including three-nucleon interactions and extensions from light toward medium-mass nuclei.

⌘

Interpretable ML for nuclear systematics

Using WINN, equivariant architectures, neural quantum states, and active learning to discover, explain, and represent emergent nuclear symmetries.

⬡

Scientific leadership & collaboration

Leading interdisciplinary projects, an equivariant-learning lecture series, and scientific programs that connect nuclear theory, AI, and computation.

Affiliations

Societies & memberships

LSU AI Journal Club Member since 2024 · Organizing Committee since 2026 Amer. Assoc. for the Advancement of Science Member · 2024 – present Sigma Xi Member · 2024 – 2025 FRIB Theory Alliance (FRIB-TA) Member · 2024 – present APS Chapter at LSU Founding member · 2024 – present American Physical Society (APS) Member · 2023 – present TalentUD, University of Debrecen Member · 2021 – 2023

Let's talk nuclei, symmetry, or AI for science.

Open to collaborations, talks, and postdoc conversations.

Get in touch