Welcome to Xin Bing's homepage

News

Nov 10, 2026	Our new paper Nonparametric Maximum Likelihood Estimation for Mixtures of Logistic Regressions is available on arXiv. It establishes nonasymptotic convergence rates for the sieve NPMLE in mixtures of logistic regressions and provides a finite-particle algorithm with provable computational guarantees.
Nov 10, 2026	Congratulations to Muheng on winning the Department of Statistical Sciences Doctoral Early Research Excellence Award!
Nov 09, 2026	Our new paper Near-Purity Screening and Dominant-Atom Estimation in Softmax Mixtures and Their Ensembles is available on arXiv. It introduces a criterion for screening near-pure softmax mixtures without knowing the number of components and establishes finite-sample guarantees for estimating dominant and shared mixture atoms.
Sep 29, 2026	Our new paper Federated Independent Component Analysis via Spectral Alignment and Robust Aggregation is available on arXiv. It combines spectral alignment with robust aggregation to estimate a shared ICA mixing matrix across heterogeneous clients, with near-minimax guarantees in the homogeneous setting.
Jul 01, 2026	Our paper Convergence and Optimality of the EM Algorithm Under Multi-Component Gaussian Mixture Models has been accepted by Biometrika, 2026+. We establish the minimax optimal rate of convergence of the EM algorithm for multi-component GMMs in full generality.
May 25, 2026	Congratulations to Bingqing on receiving the prestigious Mary H. Beatty Scholarship!
Apr 29, 2026	Congratulations to Bingqing on winning the Student/Postdoctoral Travel Scholarship for the ICSA Canada Chapter 2026 Symposium!
Apr 08, 2026	Our paper A New Regression Lens on Multi-Class Classification has been accepted by Bernoulli, 2026+. It develops a new flexible regression framework for handling multi-class classification tasks.