A Smoothing Newton Method for Rank-one Matrix Recovery
This provides a more reliable algorithm for phase retrieval, an incremental improvement in computational optimization.
The paper tackled the instability of Bures-Wasserstein gradient descent for rank-one matrix recovery in phase retrieval by developing a smoothing Newton method, achieving stable superlinear convergence in synthetic experiments.
We consider the phase retrieval problem, which involves recovering a rank-one positive semidefinite matrix from rank-one measurements. A recently proposed algorithm based on Bures-Wasserstein gradient descent (BWGD) exhibits superlinear convergence, but it is unstable, and existing theory can only prove local linear convergence for higher rank matrix recovery. We resolve this gap by revealing that BWGD implements Newton's method with a nonsmooth and nonconvex objective. We develop a smoothing framework that regularizes the objective, enabling a stable method with rigorous superlinear convergence guarantees. Experiments on synthetic data demonstrate this superior stability while maintaining fast convergence.