English

High-Dimensional Poisson DAG Model Learning Using $\ell_1$-Regularized Regression

Machine Learning 2019-05-28 v3 Machine Learning

Abstract

In this paper, we develop a new approach to learning high-dimensional Poisson directed acyclic graphical (DAG) models from only observational data without strong assumptions such as faithfulness and strong sparsity. A key component of our method is to decouple the ordering estimation or parent search where the problems can be efficiently addressed using 1\ell_1-regularized regression and the mean-variance relationship. We show that sample size n=Ω(d2log9p)n = \Omega( d^{2} \log^{9} p) is sufficient for our polynomial time Mean-variance Ratio Scoring (MRS) algorithm to recover the true directed graph, where pp is the number of nodes and dd is the maximum indegree. We verify through simulations that our algorithm is statistically consistent in the high-dimensional p>np>n setting, and performs well compared to state-of-the-art ODS, GES, and MMHC algorithms. We also demonstrate through multivariate real count data that our MRS algorithm is well-suited to estimating DAG models for multivariate count data in comparison to other methods used for discrete data.

Keywords

Cite

@article{arxiv.1810.02501,
  title  = {High-Dimensional Poisson DAG Model Learning Using $\ell_1$-Regularized Regression},
  author = {Gunwoong Park and Sion Park},
  journal= {arXiv preprint arXiv:1810.02501},
  year   = {2019}
}

Comments

43 pages

R2 v1 2026-06-23T04:29:12.414Z