Finding a large submatrix of a Gaussian random matrix
We consider the problem of finding a k × k submatrix of an n × n matrix with i.i.d. standard Gaussian entries, which has a large average entry. It was shown in [Bhamidi, Dey and Nobel (2012)] using nonconstructive methods that the largest average value of a k × k submatrix is 2(1 + o(1))√log n/k, wi...
| Principais autores: | , |
|---|---|
| Outros Autores: | |
| Formato: | Artigo |
| Publicado em: |
Institute of Mathematical Statistics
2019
|
| Acesso em linha: | http://hdl.handle.net/1721.1/120593 https://orcid.org/0000-0001-8898-8778 https://orcid.org/0000-0002-3726-1517 |