Loading...
Please wait, while we are loading the content...
Similar Documents
Robustness of SDPs for Partial Recovery of Clustering Subgaussian Mixtures UROP + Final Paper , Summer 2016
| Content Provider | Semantic Scholar |
|---|---|
| Author | Chen, Siqi |
| Copyright Year | 2016 |
| Abstract | In this paper, we examine the robustness of a relax-and-round k-means clustering procedure, a method for clustering subgaussian mixtures using semidefinite programming first introduced in [MVW16]. We are interested in the robustness of the algorithm when there is an adversarial corruption of N points each through distance at most R0. We show that under such corruption this specific algorithm well-approximates the center of the subgaussians. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://math.mit.edu/research/undergraduate/urop-plus/documents/2016/Chen.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
