Yimeng Min

Research Interests

My work sits at the intersection of AI and the natural sciences. I draw on physics, geometry, and dynamical systems to build more efficient generative models, such as flow matching, and to understand why they work. I am also interested in learning systems that are physically grounded rather than purely data-driven. On the application side, I use machine-learning-guided discovery to find new functional materials, validated experimentally.

Some current research directions include:

  • AI for Science & Scientific Machine Learning
  • Computational Sustainability
  • Generative Models
  • Geometric Deep Learning

Openings

I am recruiting several Ph.D. students to start in 2027 in Computer Science and Engineering at Texas A&M. Postdocs are welcome to reach out, as are master’s and undergraduate students. Applicants with a background in mathematics or physics are especially welcome. How to join →

Selected Publications

  1. Zhong, M.*, Tran, K.*, Min, Y.*, Wang, C.* et al. 2020. “Accelerated Discovery of CO2 Electrocatalysts Using Active Machine Learning.” Nature.
    Details

    We show that machine learning can accelerate the search. Using computer models and theoretical data, algorithms can toss out the worst options and point the way toward more promising candidates. The new catalyst we found is the first one for CO2-to-ethylene conversion to have been designed in part through the use of AI.

  2. Min, Y. and Gomes, C.P. 2026. “Learning Unbiased Permutations via Flow Matching.” NeurIPS.
  3. Min, Y. and Gomes, C.P. 2025. “Structure as Search: Unsupervised Permutation Learning for Combinatorial Optimization.” New Perspectives in Graph Machine Learning Workshop at NeurIPS.
  4. Min, Y.*, Bai, Y.*, and Gomes, C.P. 2023. “Unsupervised Learning for Solving the Travelling Salesman Problem.” NeurIPS.
  5. Min, Y.*, Wenkel, F.*, Perlmutter, M., and Wolf, G. 2022. “Can Hybrid Geometric Scattering Networks Help Solve the Maximum Clique Problem?” NeurIPS.
  6. Min, Y.*, Wenkel, F.*, and Wolf, G. 2020. “Scattering GCN: Overcoming Oversmoothness in Graph Convolutional Networks.” NeurIPS.

*Equal contribution