List-Decodable Sparse Mean Estimation

Robust mean estimation is one of the most important problems in statistics: given a set of samples in where an fraction are drawn from some distribution and the rest are adversarially corrupted, we aim to estimate the mean of . A surge of recent research interest has been focusing on the list-decodable setting where , and the goal is to output a finite number of estimates among which at least one approximates the target mean. In this paper, we consider that the underlying distribution is Gaussian with -sparse mean. Our main contribution is the first polynomial-time algorithm that enjoys sample complexity , i.e. poly-logarithmic in the dimension. One of our core algorithmic ingredients is using low-degree sparse polynomials to filter outliers, which may find more applications.
View on arXiv