Robust PCA is a widely used technique for Principal Component Analysis when the data is corrupted by outliers. The goal of the present short note is to report on the performance results of a simple modification of the method of Netrapali et al. for estimating Low Rank + Sparse models where the sparse matrix has the structure of a tree. We demonstrate the efficiency of the approach on the problem of estimating the topology in power grid networks.
|Number of pages||6|
|Journal||International Journal of Electrical Power and Energy Systems|
|Early online date||20 Mar 2018|
|Publication status||Published - 30 Sep 2018|
- non-convex optimization
- Robust PCA
- tree structured sparsity