Distributed Least Squares in Small Space via Sketching and Bias Reduction

arXiv:2405.05343v1 Announce Type: cross
Abstract: Matrix sketching is a powerful tool for reducing the size of large data matrices. Yet there are fundamental limitations to this size reduction when we want to recover an accurate estimator for a task such as least square regression. We show that these limitations can be circumvented in the distributed setting by designing sketching methods that minimize the bias of the estimator, rather than its error. In particular, we give a sparse sketching method running …

abstract arxiv bias cs.ds cs.lg cs.na data distributed estimator fundamental least limitations math.na matrix regression show small space square squares tool type via