In this talk, we will discuss manifold identification for the l1-regularization problem on the Stiefel manifold. First, we will demonstrate that the intersection of the identified manifold with the Stiefel manifold forms a submanifold. Building on this, we propose a novel second-order retraction-based algorithm specifically designed for the intersected submanifold. Numerical experiments demonstrate that the new algorithm exhibits superlinear convergence and outperform existing methods.

报告人简介：陈士祥，中国科学技术大学数学科学学院特任研究员。主要从事非凸优化算法设计和分析等方向的研究，相关成果发表在SIAM Review、 SIAM J. Opt.、IEEE TPAMI、IEEE TAC、IEEE TSP 等期刊，以及ICML、NeurIPS会议。