About me

Hi, I’m Jifan! I’m a fourth-year PhD student of Statistics and Data Science at Northwestern University. I’m advised by Prof. Miklos Racz. I received my Bachelor’s degree in Pure and applied mathematics from Tsinghua University in 2021, where I worked with Prof. Qian Lin.

Research Interest

I work at the intersection of statistics, probabilities and machine learning, with a core focus on graph-structured data, both from a theoretical and applied perspective.

On the theoretical side, I build and analyze probabilistic models of complex networks that provide rigorous guarantees for tasks such as graph matching, community detection, and latent structure recovery. I am particularly interested in leveraging multiple correlated graphs to improve the reliability and robustness of inference, an idea that naturally extends to recommender systems and networked data platforms.

On the applied side, I design machine learning algorithms for real-world graph problems. My recent work focuses on knowledge graph learning, where I investigate how pretraining can be combined with supervised learning to produce robust, high-quality embeddings for reasoning and prediction, together with provable error guarantees. I have also developed uncertainty quantification methods for spatio-temporal graphs, providing rigorous coverage guarantees in high-dimensional settings, and explored how graph neural networks can be adapted for causal representation learning from networked data.

I’m passionate about graph-structured data as a powerful abstraction that underpins a wide range of domains. I’m excited by its potential to unify theory and practice and to drive intelligent, scalable solutions across diverse applications.

News

• I am currently looking for a full-time position starting in Jun 2026, in the tech and quant domains.
  I am actively exploring roles such as Research Scientist, Data Scientist, Quantitative Researcher, or similar positions where I can contribute to impactful AI research and development.

• ✅ NeurIPS 2025 Accepted!
  My first co-authored paper, Topology-Aware Conformal Prediction on Stream Networks, has been accepted to NeurIPS 2025.

• 🗣️ INFORMS 2025 Invited Talk
  I will be giving an invited talk on the same paper at INFORMS 2025, in the session “Advances in Machine Learning and Deep Learning for Multivariate Time Series Forecasting”.

Other

In my spare time, I like playing badmintons, table tennis and I have two cats, Melon and Sherry.