Corpus ID: 229923068

The Sample Complexity of Robust Covariance Testing

  title={The Sample Complexity of Robust Covariance Testing},
  author={Ilias Diakonikolas and D. Kane},
  • Ilias Diakonikolas, D. Kane
  • Published 2020
  • Computer Science, Mathematics
  • ArXiv
  • We study the problem of testing the covariance matrix of a high-dimensional Gaussian in a robust setting, where the input distribution has been corrupted in Huber’s contamination model. Specifically, we are given i.i.d. samples from a distribution of the form Z = (1− ǫ)X + ǫB, where X is a zero-mean and unknown covariance Gaussian N (0,Σ), B is a fixed but unknown noise distribution, and ǫ > 0 is an arbitrarily small constant representing the proportion of contamination. We want to distinguish… CONTINUE READING


    Robust Estimators in High Dimensions without the Computational Intractability
    • 199
    • PDF
    Agnostic Estimation of Mean and Covariance
    • 197
    • PDF
    A New Approach for Testing Properties of Discrete Distributions
    • Ilias Diakonikolas, D. Kane
    • Mathematics, Computer Science
    • 2016 IEEE 57th Annual Symposium on Foundations of Computer Science (FOCS)
    • 2016
    • 105
    • PDF
    Statistical Query Lower Bounds for Robust Estimation of High-Dimensional Gaussians and Gaussian Mixtures
    • 108
    • Highly Influential
    • PDF
    Testing Ising Models
    • 53
    • PDF
    Sample-Optimal Identity Testing with High Probability
    • 41
    • PDF
    Optimal Algorithms and Lower Bounds for Testing Closeness of Structured Distributions
    • 39
    • PDF
    Optimal Testing for Properties of Distributions
    • 123
    • PDF