| IEICE Technical Committee Submission System Conference Paper's Information |
Online Proceedings [Sign in] Tech. Rep. Archives |
| Go Top Page | Go Previous | [Japanese] / [English] |
| Paper Abstract and Keywords | ||
| Presentation | 2015-11-27 14:00 [Poster Presentation] Non-parametric tensor learning with Gaussian process prior and its application to multi-task learning Heishiro Kanagawa, Taiji Suzuki (Titech)  IBISML2015-89 |
|
| Abstract | (in Japanese) | (See Japanese page) |
| (in English) | 低ランクテンソル推定は複数のデータソース間の高次の関係性を学習する方法として，マルチタスク学習， 推薦システム，時空間解析など様々な問題に応用されている．低ランクテンソルを推定する代表的な手法として，線 形のモデルを仮定した凸最適化に基づく手法がよく用いられる．しかし，このような線形な凸最適化手法は，データ に潜む非線形性の記述が難しいという実用的問題や，デザインへの強凸性の仮定のような強い仮定が必要であるとい う理論的問題がある．本研究では，線形モデルを再生核ヒルベルト空間での問題に拡張し，特にガウシアンプロセス 事前分布を用いたベイズ法に基づく手法を提案し，その収束レートを制限強凸性のような条件を仮定せずに示す．ま た提案手法を実データに対するマルチタスクラーニングへと適用し，既存の線形手法に比べ，高い推定精度を持つこ とを数値的に示す． |
|
| Keyword | (in Japanese) | (See Japanese page) |
| (in English) | tensor / Gaussian process / multi-task learning / / / / / | |
| Reference Info. | IEICE Tech. Rep., vol. 115, no. 323, IBISML2015-89, pp. 273-280, Nov. 2015. | |
| Paper # | IBISML2015-89 | |
| Date of Issue | 2015-11-19 (IBISML) | |
| ISSN | Print edition: ISSN 0913-5685 Online edition: ISSN 2432-6380 | |
| Copyright and reproduction |
All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) | |
| Download PDF | IBISML2015-89 |
|
| Conference Information | |
| Committee | IBISML |
| Conference Date | 2015-11-25 - 2015-11-27 |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Epochal Tsukuba |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Information-Based Induction Science Workshop (IBIS2015) |
| Paper Information | |
| Registration To | IBISML |
| Conference Code | 2015-11-IBISML |
| Language | English (Japanese title is available) |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Non-parametric tensor learning with Gaussian process prior and its application to multi-task learning |
| Sub Title (in English) | |
| Keyword(1) | tensor |
| Keyword(2) | Gaussian process |
| Keyword(3) | multi-task learning |
| Keyword(4) | |
| Keyword(5) | |
| Keyword(6) | |
| Keyword(7) | |
| Keyword(8) | |
| 1st Author's Name | Heishiro Kanagawa |
| 1st Author's Affiliation | Tokyo Institute of Technology (Titech) |
| 2nd Author's Name | Taiji Suzuki |
| 2nd Author's Affiliation | Tokyo Institute of Technology (Titech) |
| 3rd Author's Name | |
| 3rd Author's Affiliation | () |
| 4th Author's Name | |
| 4th Author's Affiliation | () |
| 5th Author's Name | |
| 5th Author's Affiliation | () |
| 6th Author's Name | |
| 6th Author's Affiliation | () |
| 7th Author's Name | |
| 7th Author's Affiliation | () |
| 8th Author's Name | |
| 8th Author's Affiliation | () |
| 9th Author's Name | |
| 9th Author's Affiliation | () |
| 10th Author's Name | |
| 10th Author's Affiliation | () |
| 11th Author's Name | |
| 11th Author's Affiliation | () |
| 12th Author's Name | |
| 12th Author's Affiliation | () |
| 13th Author's Name | |
| 13th Author's Affiliation | () |
| 14th Author's Name | |
| 14th Author's Affiliation | () |
| 15th Author's Name | |
| 15th Author's Affiliation | () |
| 16th Author's Name | |
| 16th Author's Affiliation | () |
| 17th Author's Name | |
| 17th Author's Affiliation | () |
| 18th Author's Name | |
| 18th Author's Affiliation | () |
| Speaker | Author-1 |
| Date Time | 2015-11-27 14:00:00 |
| Presentation Time | 180 minutes |
| Registration for | IBISML |
| Paper # | IBISML2015-89 |
| Volume (vol) | vol.115 |
| Number (no) | no.323 |
| Page | pp.273-280 |
| #Pages | 8 |
| Date of Issue | 2015-11-19 (IBISML) |
[Return to Top Page]