| 大会名称 |
| 2016年 ソサイエティ大会 |
| 大会コ－ド |
| 2016S |
| 開催年 |
|
2016
|
| 発行日 |
|
2016-09-06
|
| セッション番号 |
|
A-10
|
| セッション名 |
|
システム数理と応用
|
| 講演日 |
|
2016/9/20
|
| 講演場所(会議室等) |
|
工学部 情報科学研究科棟 A22
|
| 講演番号 |
|
A-10-3
|
| タイトル |
|
Riemannian gossip algorithms for decentralized matrix completion
|
| 著者名 |
|
○Bamdev Mishra, Hiroyuki Kasai, Saroop Atul,
|
| キーワード |
|
riemannian optimization, matrix completion, decentralized optimization, gossip algorithm
|
| 抄録 |
|
The problem of low-rank matrix completion amounts to completing a matrix from a small number of entries by assuming a low-rank model for the matrix. In this paper, we are propose a decentralized setting, where we divide the matrix completion problem into smaller subproblems that are solved by many agents locally while simultaneously enabling them to arrive at a consensus that solves the full problem
|
| 本文pdf |
|
PDF download
|
