Robustness meets algorithms
Author(s)
Diakonikolas, Ilias; Kamath, Gautam; Kane, Daniel M; Li, Jerry; Moitra, Ankur; Stewart, Alistair; ... Show more Show less
DownloadPublished version (2.700Mb)
Publisher with Creative Commons License
Publisher with Creative Commons License
Creative Commons Attribution
Terms of use
Metadata
Show full item recordAbstract
<jats:p>In every corner of machine learning and statistics, there is a need for estimators that work not just in an idealized model, but even when their assumptions are violated. Unfortunately, in high dimensions, being provably robust and being efficiently computable are often at odds with each other.</jats:p>
<jats:p>We give the first efficient algorithm for estimating the parameters of a high-dimensional Gaussian that is able to tolerate a constant fraction of corruptions that is independent of the dimension. Prior to our work, all known estimators either needed time exponential in the dimension to compute or could tolerate only an inverse-polynomial fraction of corruptions. Not only does our algorithm bridge the gap between robustness and algorithms, but also it turns out to be highly practical in a variety of settings.</jats:p>
Date issued
2021Department
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence LaboratoryJournal
Communications of the ACM
Publisher
Association for Computing Machinery (ACM)