Corpus ID: 5299731

The Noisy Power Method: A Meta Algorithm with Applications

@inproceedings{Hardt2014TheNP,
  title={The Noisy Power Method: A Meta Algorithm with Applications},
  author={M. Hardt and E. Price},
  booktitle={NIPS},
  year={2014}
}
  • M. Hardt, E. Price
  • Published in NIPS 2014
  • Computer Science, Mathematics
  • We provide a new robust convergence analysis of the well-known power method for computing the dominant singular vectors of a matrix that we call the noisy power method. Our result characterizes the convergence behavior of the algorithm when a significant amount noise is introduced after each matrix-vector multiplication. The noisy power method can be seen as a meta-algorithm that has recently found a number of important applications in a broad range of machine learning problems including… CONTINUE READING
    Noisy Accelerated Power Method for Eigenproblems With Applications
    An Improved Gap-Dependency Analysis of the Noisy Power Method
    30
    Accelerated Stochastic Power Iteration
    32
    Noisy Power Method with Grassmann Average
    Stochastic Variance-Reduced Heavy Ball Power Iteration.
    1
    The random component-wise power method
    3
    Non-Stationary Streaming PCA
    3
    A Framework for Private Matrix Analysis
    Distributed Stochastic Algorithms for High-rate Streaming Principal Component Analysis
    1

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 29 REFERENCES
    Understanding Alternating Minimization for Matrix Completion
    • M. Hardt
    • Computer Science, Mathematics
    • 2014
    187
    The Power of Convex Relaxation: Near-Optimal Matrix Completion
    1623
    Beating randomized response on incoherent matrices
    70
    Beyond worst-case analysis in private singular vector computation
    78
    Robust principal component analysis?
    4791
    Memory Limited, Streaming PCA
    125