A1274
Title: Partial correlation network estimation by semismooth Newton methods
Authors: Joong-Ho Won - Seoul National University (Korea, South) [presenting]
Abstract: A scalable second-order algorithm is developed for a recently proposed $\ell_1$-regularized pseudolikelihood-based partial correlation network estimation framework. While the latter method admits statistical guarantees and is inherently scalable compared to likelihood-based methods such as graphical lasso, currently available implementations rely only on first-order information and require thousands of iterations to obtain reliable estimates even on high-performance supercomputers. The inherent scalability of the framework is further investigated, and locally and globally convergent semismooth Newton methods are proposed. Despite the nonsmoothness of the problem, these second-order algorithms converge at a locally quadratic rate and require only a few tens of iterations in practice. Each iteration reduces to solving linear systems of small dimensions or linear complementary problems of smaller dimensions, making the computation suitable also for less powerful computing environments. Experiments on both simulated and real-world genomic datasets demonstrate the superior convergence behavior and computational efficiency of the proposed algorithm, which positions the method as a promising tool for massive-scale network analysis sought in modern multi-omics research.
