19
2

Nearly minimax robust estimator of the mean vector by iterative spectral dimension reduction

Abstract

We study the problem of robust estimation of the mean vector of a sub-Gaussian distribution. We introduce an estimator based on spectral dimension reduction (SDR) and establish a finite sample upper bound on its error that is minimax-optimal up to a logarithmic factor. Furthermore, we prove that the breakdown point of the SDR estimator is equal to 1/21/2, the highest possible value of the breakdown point. In addition, the SDR estimator is equivariant by similarity transforms and has low computational complexity. More precisely, in the case of nn vectors of dimension pp -- at most εn\varepsilon n out of which are adversarially corrupted -- the SDR estimator has a squared error of order (rΣn+ε2log(1/ε))logp\big(\frac{r_\Sigma}{n} + \varepsilon^2\log(1/\varepsilon)\big){\log p} and a running time of order p3+np2p^3 + n p^2. Here, rΣpr_\Sigma\le p is the effective rank of the covariance matrix of the reference distribution. Another advantage of the SDR estimator is that it does not require knowledge of the contamination rate and does not involve sample splitting. We also investigate extensions of the proposed algorithm and of the obtained results in the case of (partially) unknown covariance matrix.

View on arXiv
Comments on this paper