Robustly Learning a Gaussian: Getting Optimal Error, Efficiently
We study the fundamental problem of learning the parameters of a high-dimensional Gaussian in the presence of noise | where an "-fraction of our samples were chosen by an adversary. We give robust estimators that achieve estimation error O(ϵ) in the total variation distance, which is optimal up...
Main Authors: | Stewart, Alistair (Author), Diakonikolas, Ilias (Contributor), Kamath, Gautam Chetan (Contributor), Kane, Daniel M (Contributor), Li, Jerry Zheng (Contributor), Moitra, Ankur (Contributor) |
---|---|
Other Authors: | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory (Contributor), Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science (Contributor), Massachusetts Institute of Technology. Department of Mathematics (Contributor), Massachusetts Institute of Technology. Department of Physics (Contributor) |
Format: | Article |
Language: | English |
Published: |
Society for Industrial and Applied Mathematics,
2018-06-11T17:27:24Z.
|
Subjects: | |
Online Access: | Get fulltext |
Similar Items
-
Robust Estimators in High Dimensions without the Computational Intractability
by: Diakonikolas, Ilias, et al.
Published: (2018) -
Robust Estimators in High-Dimensions Without the Computational Intractability
by: Diakonikolas, Ilias, et al.
Published: (2020) -
Approximation of Multiobjective Optimization Problems
by: Diakonikolas, Ilias
Published: (2011) -
Provable ICA with Unknown Gaussian Noise, and Implications for Gaussian Mixtures and Autoencoders
by: Arora, Sanjeev, et al.
Published: (2017) -
Efficiently learning structured distributions from untrusted batches
by: Chen, Sitan, et al.
Published: (2021)