Efficient Algorithms and Lower Bounds for Robust Linear Regression
SODA '19: Symposium on Discrete Algorithms San Diego California January, 2019, pp. 2745-2754, 2019.
We study the prototypical problem of high-dimensional linear regression in a robust model where an ε-fraction of the samples can be adversarially corrupted. We focus on the fundamental setting where the covariates of the uncorrupted samples are drawn from a Gaussian distribution N(0, Σ) on Rd. We give nearly tight upper bounds and computa...More
Full Text (Upload PDF)
PPT (Upload PPT)