Zheng Li

AI · Causal Discovery and Inference

Zheng Li 李政

I am currently a Ph.D. student at Fudan University (2026–present), under the supervision of Prof. Shuigeng Zhou.

I was a Research Assistant at the Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences (2025–2026), under the supervision of Prof. Hao Zhang and Prof. Ruxin Wang.

I received my M.S. degree from Beijing Technology and Business University (September 2023–July 2025), under the supervision of Prof. Feng Xie. During this time, I was also a student member of the Causal Inference Group led by Prof. Zhi Geng.

Research interests: causal discovery in the presence of latent variables and selection bias; selecting valid adjustment sets for causal effect estimation; testability of instrumental variables.

I welcome academic discussions and collaborations. Please feel free to contact me.

Selected Work

Publications

* Equal contribution.

Preprints

  1. Local Causal Structure Learning in the Presence of Latent Variables and Selection Bias

    Zheng Li, Hao Zhang, Ruxin Wang, Ruichu Cai, Kun Zhang, Feng Xie.

    arXiv:2607.19866 PDF Code

Journal

  1. Testability of Instrumental Variables in Additive Nonlinear, Non-Constant Effects Models

    Xichen Guo*, Zheng Li*, Biwei Huang, Yan Zeng, Zhi Geng, Feng Xie.

    Journal of Machine Learning Research, 2026. PDF Code

Conference

  1. A Recursive Decomposition Framework for Causal Structure Learning in the Presence of Latent Variables

    Zheng Li, Feng Xie, Shenglan Nie, Xichen Guo, Ruxin Wang, Hao Zhang.

    ICML, Seoul, South Korea, 2026. Oral Top 0.71% of 23,918 PDF Code

  2. Local Learning for Covariate Selection in Nonparametric Causal Effect Estimation with Latent Variables

    Zheng Li, Xichen Guo, Feng Xie, Yan Zeng, Hao Zhang, Zhi Geng.

    NeurIPS, San Diego, USA, 2025. PDF Code

  3. Local Identifying Causal Relations in the Presence of Latent Variables

    Zheng Li, Zeyu Liu, Feng Xie, Hao Zhang, Chunchen Liu, Zhi Geng.

    ICML, Vancouver, Canada, 2025. 🛠️ Spotlight Top 2.6% of 12,107

  4. Local Causal Structure Learning in the Presence of Latent Variables

    Feng Xie, Zheng Li, Peng Wu, Yan Zeng, Chunchen Liu, Zhi Geng.

    ICML, Vienna, Austria, 2024. PDF Code

Software

Open Source Projects

FastDAG2PAG: DAG to PAG/MAG Converter

A Python implementation for transforming directed acyclic graphs (DAGs) into maximal ancestral graphs (MAGs) and partial ancestral graphs (PAGs), given specified latent variables and selection bias. The tool implements Theorem 4.2 from Richardson and Spirtes, "Ancestral Graph Markov Models" (The Annals of Statistics, 2002).

Community

Academic Service

  • Conference Program Committee or Reviewer: ICML 2026; NeurIPS 2026.
  • Journal Reviewer: Transactions on Machine Learning Research (TMLR).

Recognition

Awards & Honors