I am currently a postdoctoral scholar at Department of Computational Medicine, UCLA, working with Dr. Kenneth Lange and Dr. Hua Zhou. I received my Ph.D. from North Carolina State University, where I primarily worked with Dr. Eric C. Chi (currently at Rice) on numerical optimization in statistical learning problems.

📝 Preprints & In Preparation

Minimum Covariance Determinant: Spectral Embedding and Subset Size Determination

Qiang Heng, Hui Shen, Kenneth Lange

Submitted

• A spectral approach to the minimum covariance determinant problem, coupled with a novel bootstrap procedure for estimating the number of inliers/outliers. [arXiv]

Trimmed GLM Estimation via Variance Stabilized Transformation

Qiang Heng, Kenneth Lange

Working paper

Anderson Accelerated Operator Splitting Methods for Convex-nonconvex Regularized Problems

Qiang Heng, Xiaoqian Liu, Shiqian Ma, Eric C. Chi

To be submitted

• Globally convergent type-II Anderson acceleration for several common splitting methods, with application in convex-nonconvex regularization.

📝 Publications With Primary Role

Technometrics

Robust Low-rank Tensor Decomposition with the L2 Criterion

Qiang Heng, Eric C. Chi, Yufeng Liu

Technometrics, 2023

• A nonconvex, differentiable formulation of robust Tucker decomposition that exhibits stronger recovery capabilities in more challenging high-rank scenarios. [paper][code][arXiv]

JCGS

Bayesian Trend Filtering via Proximal Markov Chain Monte Carlo

Qiang Heng, Hua Zhou, Eric C. Chi

Journal of Computational and Graphical Statistics, 2023

• A novel MCMC methodology which uses epigrahph priors and Moreau envelopes to sample from nonsmooth posterior density functions with applications in Bayesian trend filtering. [paper][code][arXiv]

📝 Publications with Supporting Role

• FreeMatch: Self-adaptive Thresholding for Semi-supervised Learning [arXiv] Yidong Wang, Hao Chen, Qiang Heng, Wenxin Hou, Yue Fan, Zhen Wu, Jindong Wang, Marios Savvides, Takahiro Shinozaki, Bhiksha Raj, Bernt Schiele, Xing Xie. The Eleventh International Conference on Learning Representations (ICLR 2023).

• Proximal MCMC for Bayesian Inference of Constrained and Regularized Estimation [arXiv] Xinkai Zhou, Qiang Heng, Eric C. Chi, Hua Zhou. The American Statistician (accepted, 2024+).

📖 Educations

• 2019.08 - 2023.12, Ph.D. Statistics, North Carolina State University.
• 2015.09 - 2019.06, B.S. Statistics, Shanghai University of Finance and Economics.

💬 Presentations

• 2024.04 Nanjing Audit University, Minimum Covariance Determinant: Spectral Embedding and Subset Size Determination, Invited Talk, Seminar
• 2024.03 McGill University, Minimum Covariance Determinant: Spectral Embedding and Subset Size Determination, Invited Talk, Statistics Seminar
• 2023.07 Brown University, Anderson Accelerated Operator Splitting for Convex-nonconvex Regularization, Poster Presentation, Acceleration and Extrapolation Methods workshop at ICERM
• 2022.07 Lehigh University, Bayesian Trend Filtering via Proximal Markov Chain Monte Carlo, Invited Talk, ICCOPT 2022