講演抄録/キーワード |
講演名 |
2015-11-26 15:00
[ポスター講演]Learning Structure of Partial Markov Random Field via Partitioned Ratio ○Song Liu(ISM)・Taiji Suzuki(Tokyo Tech.)・Masashi Sugiyama(UTokyo)・Kenji Fukumizu(ISM) IBISML2015-72 |
抄録 |
(和) |
A new concept, partitioned ratio is proposed to find the partial connectivity of the Markov random field. First we argue this partitioned ratio has a profound link with the Markov properties of random variables via its factorization. Specifically, partitioned ratio may be further decomposed into Bridges, a novel subgraph structure, capturing the partial connectivity of the Markov random field, which can be roughly considered as the ``link'' structure between two partitions. Second, a simple one-shot optimization is illustrated to learn the sparse factorizations of partitioned ratio efficiently, regardless the Gaussianity of the joint distribution or the marginal distributions. |
(英) |
A new concept, partitioned ratio is proposed to find the partial connectivity of the Markov random field. First we argue this partitioned ratio has a profound link with the Markov properties of random variables via its factorization. Specifically, partitioned ratio may be further decomposed into Bridges, a novel subgraph structure, capturing the partial connectivity of the Markov random field, which can be roughly considered as the ``link'' structure between two partitions. Second, a simple one-shot optimization is illustrated to learn the sparse factorizations of partitioned ratio efficiently, regardless the Gaussianity of the joint distribution or the marginal distributions. |
キーワード |
(和) |
Markov Random Field / Structure Learning / Density Ratio Estimation / / / / / |
(英) |
Markov Random Field / Structure Learning / Density Ratio Estimation / / / / / |
文献情報 |
信学技報, vol. 115, no. 323, IBISML2015-72, pp. 147-151, 2015年11月. |
資料番号 |
IBISML2015-72 |
発行日 |
2015-11-19 (IBISML) |
ISSN |
Print edition: ISSN 0913-5685 Online edition: ISSN 2432-6380 |
著作権に ついて |
技術研究報告に掲載された論文の著作権は電子情報通信学会に帰属します.(許諾番号:10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
PDFダウンロード |
IBISML2015-72 |