Communication-Efficient Distributed Eigenspace Estimation with Arbitrary Node Failures

Abstract

We develop an eigenspace estimation algorithm for distributed environments with arbitrary node failures, where a subset of computing nodes can return structurally valid but otherwise arbitrarily chosen responses. Notably, this setting encompasses several important scenarios that arise in distributed computing and data-collection environments such as silent/soft errors, outliers or corrupted data at certain nodes, and adversarial responses. Our estimator builds upon and matches the performance of a recently proposed non-robust estimator up to an additive $\tilde{O}(\sigma \sqrt{\alpha})$ error, where $\sigma^2$ is the variance of the existing estimator and $\alpha$ is the fraction of corrupted nodes.

Cite

Text

Charisopoulos and Damle. "Communication-Efficient Distributed Eigenspace Estimation with Arbitrary Node Failures." Neural Information Processing Systems, 2022.

Markdown

[Charisopoulos and Damle. "Communication-Efficient Distributed Eigenspace Estimation with Arbitrary Node Failures." Neural Information Processing Systems, 2022.](https://mlanthology.org/neurips/2022/charisopoulos2022neurips-communicationefficient/)

BibTeX

@inproceedings{charisopoulos2022neurips-communicationefficient,
  title     = {{Communication-Efficient Distributed Eigenspace Estimation with Arbitrary Node Failures}},
  author    = {Charisopoulos, Vasileios and Damle, Anil},
  booktitle = {Neural Information Processing Systems},
  year      = {2022},
  url       = {https://mlanthology.org/neurips/2022/charisopoulos2022neurips-communicationefficient/}
}