A variational approach to stable principal component pursuit
This work addresses signal processing and data analysis challenges for researchers and practitioners, but it is incremental as it builds on existing SPCP methods.
The authors tackled the problem of decomposing noisy signals into low-rank and sparse representations by introducing a new convex formulation for stable principal component pursuit (SPCP), resulting in improved scalability and practical parameter selection compared to classical SPCP formulations.
We introduce a new convex formulation for stable principal component pursuit (SPCP) to decompose noisy signals into low-rank and sparse representations. For numerical solutions of our SPCP formulation, we first develop a convex variational framework and then accelerate it with quasi-Newton methods. We show, via synthetic and real data experiments, that our approach offers advantages over the classical SPCP formulations in scalability and practical parameter selection.