Reproduce “Polylogarithmic Sparsity of Randomly Reweighted NPMLEs for Gaussian Mixtures” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Hansheng Jiang
The paper introduces a randomly reweighted nonparametric maximum-likelihood estimator for Gaussian location mixtures. It maximizes likelihood with mean-one Gamma weights concentrated near one, without prescribing component count. Under fixed-dimension, compact-support, and correct-specification assumptions, the exact estimator itself is unique and has polylogarithmic support, with likelihood and density-risk guarantees; these do not apply to numerical approximations’ atom counts.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.