Yinuo Ren

PhD Candidate
Institute for Computational and Mathematical Engineering (ICME)
Stanford University

Email: yinuoren (at) stanford (dot) edu

Google Scholar      Semantic Scholar

I am a PhD candidate in Computational and Mathematical Engineering at Institute for Computational and Mathematical Engineering (ICME), Stanford University. I am privileged to be supervised by Prof. Lexing Ying (Mathematics, Stanford) and Prof. Grant M. Rotskoff (Theoretical Chemistry, Stanford). My research interests lie in the intersection of scientific machine learning, neural network, stochastic simulation, numerical analysis, and statistics.

During summer 2023, I was an Applied Scientist intern at Amazon, working in multi-objective optimization (our AISTATS paper).

I obtained my BSc in Computational Mathematics from School of Mathematical Sciences, Peking University, supervised by Prof. Ruo Li.

Publications

  • Yinuo Ren, Chao Ma, and Lexing Ying. Understanding the Generalization Benifits of Late Learning Rate Decay. International Conference on Artificial Intelligence and Statistics (AISTATS). PMLR, 2024.  arXiv

  • Yinuo Ren, Tesi Xiao, Tanmay Gangwani, Anshuka Rangi, Holakou Rahmanian, Lexing Ying, and Subhajit Sanyal. Multi-Objective Optimization via Wasserstein-Fisher-Rao Gradient Flow.International Conference on Artificial Intelligence and Statistics (AISTATS). PMLR, 2024.  arXiv

  • Yinuo Ren, Yiping Lu, Lexing Ying, and Grant M. Rotskoff. Statistical Spatially-inhomogeneous Diffusion Inference. Proceedings of the AAAI Conference on Artificial Intelligence, 2024.  arXiv  Poster

  • Yinuo Ren, Hongli Zhao, Yuehaw Khoo, and Lexing Ying. High-dimensional Density Estimation with Tensorizing Flow. Research in the Mathematical Sciences, 2023.  arXiv

  • Haoya Li, Yuehaw Khoo, Yinuo Ren, Lexing Ying. Solving for high dimensional committor functions using neural network with online approximation to derivatives. Mathematical and Scientific Machine Learning. PMLR, 2021.  arXiv

  • Ruo Li, Yinuo Ren, Yanli Wang. Hermite spectral method for Fokker-Planck-Landau equation modeling collisional plasma. Journal of Computational Physics, 2021.  arXiv

Talks

  • High-dimensional Density Estimation with Tensorizing Flow, Tensor Network Reading Group, Université de Montréal and Mila - Quebec AI Institute, Quebec, Canada, Nov. 2023

  • Hermite spectral method for Fokker-Planck-Landau equation modeling collisional plasma, Minisymposium on Efficient Numerical Methods for Kinetic Equations and Their Applications, the 18th Annual Meeting of China Society for Industrial and Applied Mathematics, Hunan, China, Oct. 2020

  • Numerical Simulation of Plasma Instabilities Using Hermite-Galerkin Spectral Method, Workshop on Modeling, Algorithm and Analysis on Complex Fluid Dynamics, Beijing Computational Science Research Center, Beijing, China, Nov. 2019

Services

I served as a reviewer for Journal of Scientific Computing, Kinetic and Related Models, MSML 2022, ICML 2023, NeurIPS 2023, ICLR 2024, AISTATS 2024, ICML 2024.