Title:
Robust Mean Estimation in Nearly-Linear Time
Robust Mean Estimation in Nearly-Linear Time
dc.contributor.author | Hopkins, Samuel | |
dc.contributor.corporatename | Georgia Institute of Technology. Algorithms, Randomness and Complexity Center | en_US |
dc.contributor.corporatename | University of California, Berkeley. Dept. of Electrical Engineering and Computer Sciences | en_US |
dc.date.accessioned | 2019-12-09T18:04:23Z | |
dc.date.available | 2019-12-09T18:04:23Z | |
dc.date.issued | 2019-12-02 | |
dc.description | Presented on December 2, 2019 at 11:00 a.m. in the Klaus Advanced Computing Building, Room 1116E. | en_US |
dc.description | Samuel Hopkins is a Miller fellow in the Theory Group in the Electrical Engineering and Computer Sciences Department at the University of California, Berkeley. His research interests include algorithms, theoretical machine learning, semidefinite programming, sum of squares optimization, convex hierarchies, and hardness of approximation. | en_US |
dc.description | Runtime: 56:43 minutes | en_US |
dc.description.abstract | Robust mean estimation is the following basic estimation question: given i.i.d. copies of a random vector X in d-dimensional Euclidean space of which a small constant fraction are corrupted, how well can you estimate the mean of the distribution? This is a classical problem in statistics, going back to the 60's and 70's, and has recently found application to many problems in reliable machine learning. However, in high dimensions, classical algorithms for this problem either were (1) computationally intractable, or (2) lost poly(d) factors in their accuracy guarantees. Recently, polynomial time algorithms have been demonstrated for this problem that still achieve (nearly) optimal error guarantees. However, the running times of these algorithms were either at least quadratic in dimension or in 1/(desired accuracy), running time overhead which renders them ineffective in practice. In this talk we give the first truly nearly linear time algorithm for robust mean estimation which achieves nearly optimal statistical performance. Our algorithm is based on the matrix multiplicative weights method. Based on joint work with Yihe Dong and Jerry Li, to appear in NeurIPS 2019. | en_US |
dc.format.extent | 56:43 minutes | |
dc.identifier.uri | http://hdl.handle.net/1853/62093 | |
dc.language.iso | en_US | en_US |
dc.relation.ispartofseries | Algorithms and Randomness Center (ARC) Colloquium | |
dc.subject | Robust Mean Estimation | en_US |
dc.subject | Time | en_US |
dc.title | Robust Mean Estimation in Nearly-Linear Time | en_US |
dc.type | Moving Image | |
dc.type.genre | Lecture | |
dspace.entity.type | Publication | |
local.contributor.corporatename | Algorithms and Randomness Center | |
local.contributor.corporatename | College of Computing | |
local.relation.ispartofseries | ARC Colloquium | |
relation.isOrgUnitOfPublication | b53238c2-abff-4a83-89ff-3e7b4e7cba3d | |
relation.isOrgUnitOfPublication | c8892b3c-8db6-4b7b-a33a-1b67f7db2021 | |
relation.isSeriesOfPublication | c933e0bc-0cb1-4791-abb4-ed23c5b3be7e |
Files
Original bundle
1 - 4 of 4
No Thumbnail Available
- Name:
- hopkins.mp4
- Size:
- 454.34 MB
- Format:
- MP4 Video file
- Description:
- Download video
No Thumbnail Available
- Name:
- hopkins_videostream.html
- Size:
- 1.32 KB
- Format:
- Hypertext Markup Language
- Description:
- Streaming video
No Thumbnail Available
- Name:
- transcript.txt
- Size:
- 52.72 KB
- Format:
- Plain Text
- Description:
- Transcription
- Name:
- thumbnail.jpg
- Size:
- 61.63 KB
- Format:
- Joint Photographic Experts Group/JPEG File Interchange Format (JFIF)
- Description:
- Thumbnail
License bundle
1 - 1 of 1
No Thumbnail Available
- Name:
- license.txt
- Size:
- 3.13 KB
- Format:
- Item-specific license agreed upon to submission
- Description: