ダウンロード数: 86
このアイテムのファイル:
ファイル | 記述 | サイズ | フォーマット | |
---|---|---|---|---|
jmlr.proc_v51.pdf | 3.89 MB | Adobe PDF | 見る/開く |
タイトル: | New Resistance Distances with Global Information on Large Graphs. |
著者: | Nguyen, Canh Hao Mamitsuka, Hiroshi https://orcid.org/0000-0002-6607-5617 (unconfirmed) |
著者名の別形: | 馬見塚, 拓 |
発行日: | May-2016 |
出版者: | Microtome Publishing Microtome Publishing |
誌名: | JMLR Workshop and Conference Proceedings |
巻: | 51 |
開始ページ: | 639 |
終了ページ: | 647 |
抄録: | We consider the problem that on large random geometric graphs, random walk-based distances between nodes do not carry global information such as cluster structure. Instead, as the graphs become larger, the distances contain mainly the obsolete information of local density of the nodes. Many distances or similarity measures between nodes on a graph have been proposed but none are both proved to overcome this problem or computationally feasible even for small graphs. We propose new distance functions between nodes for this problem. The idea is to use electrical flows with different energy functions. Our proposed distances are proved analytically to be metrics in L^p spaces, to keep global information, avoiding the problem, and can be computed efficiently for large graphs. Our experiments with synthetic and real data confirmed the theoretical properties and practical performances of our proposed distances. |
記述: | Appearing in Proceedings of the 19th International Conference on Artificial Intelligence and Statistics (AISTATS) 2016, Cadiz, Spain. JMLR: W&CP volume 51. |
著作権等: | Copyright 2016 by the authors. |
URI: | http://hdl.handle.net/2433/219138 |
関連リンク: | http://www.jmlr.org/proceedings/papers/v51/nguyen16a.html |
出現コレクション: | 学術雑誌掲載論文等 |
このリポジトリに保管されているアイテムはすべて著作権により保護されています。