Oral
Alternating Minimizations Converge to Second-Order Optimal Solutions
Qiuwei Li · Zhihui Zhu · Gongguo Tang

Wed Jun 12th 03:15 -- 03:20 PM @ Room 104
in Optimization: Convex and Non-convex »

This work studies the second-order convergence for both standard alternating minimization and proximal alternating minimization. We show that under mild assumptions on the (nonconvex) objective function, both algorithms avoid strict saddles almost surely from random initialization. Together with known first-order convergence results, this implies both algorithms converge to a second-order stationary point. This solves an open problem for the second-order convergence of alternating minimization algorithms that have been widely used in practice to solve large-scale nonconvex problems due to their simple implementation, fast convergence, and superb empirical performance.

Author Information

Qiuwei Li (Colorado School of Mines)
Zhihui Zhu (Johns Hopkins University)
Gongguo Tang (Colorado School of Mines)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors