Hello! I’m Haolin, a Postdoctoral Researcher at Department of Statistics and Data Science (STA) at CUHK, under supervision of Prof. Junhui Wang. I obtained my Ph.D. from the Department of Statistics at Columbia University and was advised by Prof. Arian Maleki and Prof. Victor de la Peña. Before my doctoral study, I received my Master in Actuarial Science from Columbia University in 2020. I received my Bachelor in Applied Mathematics and Bachelor in Economics from Peking University in 2018. It’s a long and fruitful journey of education indeed!

My Research Directions

I have a broad interest in using statistical theory to understand current machine learning methods and developing new algorithms, with a primary focus on proportional high dimensional regime. My current research focuses on:

1. Risk estimation and cross validation methods,
2. Model adaptation and machine unlearning,
3. Data attribution and interpretability,
4. Network analysis.

Further Interests

1. Model memory and continual learning. Afterall, how can we study forgetting if we don’t understand memory?
2. Integration and symbiosis of human and AI.

Selected Publications

Publications

1. Tong, H., Ghosh, S., Zou, H. and Maleki, A. (2026). Imperfect Influence, Preserved Rankings: A Theory of TRAK for Data Attribution. Accepted at NeurIPS 2026.
2. Zou, H., Yao, H. and de la Peña, V. (2026). Exact Moment Formulae for Mean-normalized Statistics with Applications to Inequality Measures. Statistics and Probability Letters, 236:110754.
3. Pandey, A., Auddy, A., Zou, H., Maleki, A. and Kulkarni, S. (2025). Gaussian Certified Unlearning in High Dimensions: A Hypothesis Testing Approach. ICLR 2026 with oral presentation (top 4%).
4. Zou, H., Auddy, A., Kwon, Y., Rahnama Rad, K. and Maleki, A. (2025). Certified Machine Unlearning Under High Dimensional Regime. Journal of Machine Learning Research (JMLR), 26(308):1-58.
5. Zou, H. et al. (2025). Newfluence: Boosting Model Interpretability and Understanding in High Dimensions. ICML 2025, Workshop: Assessing World Models: Methods and Metrics for Evaluating Understanding.
6. Zou, H. et al. (2025). Theoretical Analysis of Leave-one-out Cross Validation for Non-differentiable Penalties under High-dimensional Settings._ Proceedings of The 28th International Conference on Artificial Intelligence and Statistics_, PMLR 258:4033-4041.
7. Auddy, A., Zou, H., Rahnama Rad, K. and Maleki, A. (2024) Approximate Leave-one-out Cross Validation for Regression with L1 Regularizers. IEEE Transactions on Information Theory, 70(11):8040–8071.
8. Auddy, A., Zou, H., Rahnama Rad, K. and Maleki, A. (2024) Approximate Leave-one-out Cross Validation for Regression with L1 Regularizers. Proceedings of The 27th International Conference on Artificial Intelligence and Statistics, PMLR 238:2377-2385. Selected for oral presentation.
9. Baydil, B., de la Peña, V., Zou, H. and Yao, H. (2025). Unbiased estimation of the Gini coefficient. Statistics and Probability Letters, 222:110376.
10. de la Peña, V., Gzyl, H., Mayoral, S., Zou, H. and Alemayehu, D. (2024). Prediction and estimation of random variables with infinite mean or variance. Communications in Statistics - Theory and Methods, 54(1):1-15.

Working Papers

1. Zou, H., Auddy, A., Rahnama Rad, K. and Maleki, A. (2026). The Bias and Variance of K-fold Cross Validation Under High-dimensional Settings. Preparing for submission.
2. Shang, C., Zou, H., Zhu, B., and Maleki, A. (2026). Error Accumulation in Second Order Sequential Machine Unlearning. Preparing for submission.
3. Zou, H. and Wang, J. (2026). Dyad and Node Attribution for Statistical Network Models. Ongoing Project.

Under Review