Pr. Xie’s research lies at the intersection of statistics, optimization, and machine learning. I work on developing computationally efficient and statistically powerful algorithms with guarantees, for engineering problems arising from various real-world applications. My work has generated societal and policy impact: see here for our work on data-driven policing.
Sequential data: Point processes, (high-dimensional and non-stationary) time series, spatio-temporal data, dynamic networks, and sequential prediction.
Hypothesis tests: Change-point detection, distributions shift, anomaly detection, two-sample test, robust hypothesis test.
Inference on structured data: Sparsity, low-rank matrix/tensor, and manifold data.
Machine learning: Uncertainty quantification, distributional robust optimization, neural networks.
Applications: Policing/crime data, power systems, health-care and medical data, material sciences, sensor networks, and logistic networks.