Poster
A Unified Variance Reduction-Based Framework for Nonconvex Low-Rank Matrix Recovery
Lingxiao Wang · Xiao Zhang · Quanquan Gu

Tue Aug 8th 06:30 -- 10:00 PM @ Gallery #20

We propose a generic framework based on a new stochastic variance-reduced gradient descent algorithm for accelerating nonconvex low-rank matrix recovery. Starting from an appropriate initial estimator, our proposed algorithm performs projected gradient descent based on a novel semi-stochastic gradient specifically designed for low-rank matrix recovery. Based upon the mild restricted strong convexity and smoothness conditions, we derive a projected notion of the restricted Lipschitz continuous gradient property, and prove that our algorithm enjoys linear convergence rate to the unknown low-rank matrix with an improved computational complexity. Moreover, our algorithm can be employed to both noiseless and noisy observations, where the (near) optimal sample complexity and statistical rate can be attained respectively. We further illustrate the superiority of our generic framework through several specific examples, both theoretically and experimentally.

Author Information

Lingxiao Wang (University of Virginia)
Xiao Zhang (University of Virginia)
Quanquan Gu (University of Virginia)

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

More from the Same Authors