Optimal convex lifted sparse phase retrieval and PCA with an atomic matrix norm regularizer
arXiv:2111.04652 · doi:10.1109/TIT.2022.3228508
Abstract
We present novel analysis and algorithms for solving sparse phase retrieval and sparse principal component analysis (PCA) with convex lifted matrix formulations. The key innovation is a new mixed atomic matrix norm that, when used as regularization, promotes low-rank matrices with sparse factors. We show that convex programs with this atomic norm as a regularizer provide near-optimal sample complexity and error rate guarantees for sparse phase retrieval and sparse PCA. While we do not know how to solve the convex programs exactly with an efficient algorithm, for the phase retrieval case we carefully analyze the program and its dual and thereby derive a practical heuristic algorithm. We show empirically that this practical algorithm performs similarly to existing state-of-the-art algorithms.
References in corpus (9)
- Phase Retrieval via Wirtinger Flow: Theory and Algorithms
- The Convex Geometry of Linear Inverse Problems
- Sparse PCA: Optimal rates and adaptive estimation
- Compressive Phase Retrieval via Generalized Approximate Message Passing
- Compressive Phase Retrieval From Squared Output Measurements Via Semidefinite Programming
- Statistical and computational trade-offs in estimation of sparse principal components
- Tight convex relaxations for sparse matrix factorization
- Simultaneous structures in convex signal recovery - revisiting the convex combination of norms
- Lifting high-dimensional nonlinear models with Gaussian regressors