Semester of Graduation
Electrical and Computer Engineering
First Major Professor
Master of Science (MS)
We study the problem of subspace tracking (ST) in the presence of missing and corrupted data. We are able to show that, under assumptions on only the algorithm inputs (input data and/or initialization), the output subspace estimates are close to the true data subspaces at all times. The guarantees hold under mild and easily interpretable assumptions and handle time-varying subspaces. We also show that our algorithm and its extensions are fast and have competitive experimental performance when compared with existing methods. Finally, this solution can be interpreted as a provably correct mini-batch and memory-efficient solution to low rank Matrix Completion (MC).
Daneshpajooh, Vahid, "Subspace tracking from missing and corrupted data using NORST and its heuristic extensions" (2019). Creative Components. 303.