Committee |
Date Time |
Place |
Paper Title / Authors |
Abstract |
Paper # |
IBISML |
2023-12-21 15:00 |
Tokyo |
National Institute of Informatics (Primary: On-site, Secondary: Online) |
Causal Effect Estimation on Hierarchical Spatial Graph Data Koh Takeuchi (Kyoto Univ.), Ryo Nishida (AIST), Hisashi Kashima (Kyoto Univ.), Masaki Onishi (AIST) IBISML2023-37 |
[more] |
IBISML2023-37 pp.42-49 |
PRMU, IBISML, IPSJ-CVIM [detail] |
2023-03-02 09:00 |
Hokkaido |
Future University Hakodate (Primary: On-site, Secondary: Online) |
Toward Regularizing Neural Networks with Meta-Learning Generative Models Shin'ya Yamaguchi (NTT/Kyoto Univ.), Daiki Chijiwa, Sekitoshi Kanai, Atsutoshi Kumagai (NTT), Hisashi Kashima (Kyoto Univ.) PRMU2022-58 IBISML2022-65 |
This paper investigates methods for improving generative data augmentation for deep learning. Generative data augmentati... [more] |
PRMU2022-58 IBISML2022-65 pp.1-6 |
IBISML |
2020-01-09 14:55 |
Tokyo |
ISM |
Approximation Ratios of Graph Neural Networks for Combinatorial Problems Ryoma Sato, Makoto Yamada, Hisashi Kashima (Kyoto Univ.) IBISML2019-22 |
[more] |
IBISML2019-22 pp.31-38 |
DE, IPSJ-DBS, IPSJ-IFAT |
2017-09-18 14:50 |
Tokyo |
Ochanomizu University |
Cost-sensitive no-show prediction for airline companies Yuji Horiguchi, Yukino Baba, Hisashi Kashima (Kyoto Univ.), Takeshi Kojima, Hiroki Kayahara, Jun Maeno (Peach Aviation) DE2017-17 |
In this research, we estimate the number of passengers (NO--SHOW) who do not appear at the airport on the day of departu... [more] |
DE2017-17 pp.57-62 |
IBISML |
2016-11-17 14:00 |
Kyoto |
Kyoto Univ. |
Budgeted stream-based active learning via adaptive submodular maximization Kaito Fujii, Hisashi Kashima (Kyoto Univ.) IBISML2016-74 |
Active learning enables us to reduce the annotation cost by adaptively selecting unlabeled instances to be labeled. For ... [more] |
IBISML2016-74 pp.199-206 |
IBISML |
2016-11-17 17:00 |
Kyoto |
Kyoto Univ. |
A Generalized Model for Multidimensional Intransitivity Jiuding Duan, Jiyi Li, Yukino Baba, Hisashi Kashima (Kyoto Univ.) IBISML2016-98 |
Recent data-driven approach for pairwise preference modelling reveals a rational way to model the intransitivity. Intran... [more] |
IBISML2016-98 pp.375-380 |
NC, IPSJ-BIO, IBISML, IPSJ-MPS (Joint) [detail] |
2015-06-23 16:10 |
Okinawa |
Okinawa Institute of Science and Technology |
Optimal Algorithms in Dueling Bandit Problem Junpei Komiyama, Junya Honda (U-Tokyo), Hisashi Kashima (Kyoto University), Hiroshi Nakagawa (U-Tokyo) IBISML2015-14 |
We study the K-armed dueling bandit problem, a variation of the standard stochastic bandit problem where the feedback is... [more] |
IBISML2015-14 pp.87-94 |
IBISML |
2015-03-06 10:30 |
Kyoto |
Kyoto University |
Quality control in human-machine hybrid crowdsourcing Toshihiro Watanabe (UTokyo), Toshinari Itoko, Shin Saito, Masatomo Kobayashi, Hironobu Takagi (IBM), Hisashi Kashima (Kyoto Univ.) IBISML2014-91 |
The power of crowdsourcing has dramatically shortened the required time to create accessible content for disabled people... [more] |
IBISML2014-91 pp.47-54 |
PRMU, IBISML, IPSJ-CVIM [detail] |
2014-09-02 15:15 |
Ibaraki |
|
On Improvement of Tensor Spectral Method for Estimation of Mixed Membership Stochastic Block Model Wataru Kaigaishi (Univ. of Tokyo), Hisashi Kashima (Kyoto Univ.) PRMU2014-51 IBISML2014-32 |
Maximum likelihood estimation is a common approach to estimating mixed membership stochastic block models; however, it i... [more] |
PRMU2014-51 IBISML2014-32 pp.131-136 |
AI, SC |
2014-08-20 14:00 |
Tokyo |
National Institute of Informatics |
Making Legacy Open Data Machine Readable by Crowdsourcing Satoshi Oyama (Hokkaido Univ.), Yukino Baba, Ikki Ohmukai (NII), Hiroaki Dokoshi (Hokkaido Univ.), Hisashi Kashima (Kyoto Univ.) AI2014-11 SC2014-8 |
[more] |
AI2014-11 SC2014-8 pp.1-6 |
IBISML |
2014-03-06 13:25 |
Nara |
Nara Women's University |
Simultaneous prediction of multiple physical properties using multi-task learning Tomoaki Iwase (Univ. of Tokyo), Atsuto Seko (Kyoto Univ.), Hisashi Kashima (Univ. of Tokyo) IBISML2013-68 |
We apply several existing techniques and a new model of multi-task learning to the problem of predicting multiple physic... [more] |
IBISML2013-68 pp.9-13 |
IBISML |
2014-03-06 15:20 |
Nara |
Nara Women's University |
Focused Tensor Completion
-- Transfer learning for completing a certain slice of a third-order tensor Data -- Taketo Akama, Yukino Baba, Hisashi Kashima (Univ. of Tokyo) IBISML2013-72 |
[more] |
IBISML2013-72 pp.39-46 |
IBISML |
2014-03-07 09:40 |
Nara |
Nara Women's University |
A Label Completion Approach to Crowd Approximation Toshihiro Watanabe, Hisashi Kashima (Univ. of Tokyo) IBISML2013-73 |
Majority vote is one of the most common methods for crowdsourced label aggregation to get higher-quality labels, but it ... [more] |
IBISML2013-73 pp.47-53 |
IBISML |
2013-07-18 13:25 |
Tokyo |
Nishiwaseda Campus (Waseda univ.) |
A Simultaneous Completion Method for Multiple Relational Data Yutaka Ieiri, Hisashi Kashima (Univ. Tokyo) IBISML2013-6 |
In this paper, we consider a completion problem of
multiple relational data sets with missing values.
In cases where ... [more] |
IBISML2013-6 pp.35-41 |
IBISML |
2012-03-13 14:40 |
Tokyo |
The Institute of Statistical Mathematics |
Matrix and Tensor Factorization with Aggregated Observations Yoshifumi Aimoto, Hisashi Kashima (Univ. of Tokyo) IBISML2011-103 |
Matrix and tensor factorization with low-rank assumption are fundamental tools in data analysis. However, the existing m... [more] |
IBISML2011-103 pp.109-116 |
IBISML |
2011-11-09 15:45 |
Nara |
Nara Womens Univ. |
A kernel-based approach for matrix and tensor completion Kohei Hayashi, Takashi Takenouchi (NAIST), Ryota Tomioka, Hisashi Kashima (Univ. Tokyo) IBISML2011-53 |
We study a new kernel-based framework for matrix and tensor completion problems. Our model provides a consistent way to ... [more] |
IBISML2011-53 pp.71-77 |
IBISML |
2011-11-10 15:45 |
Nara |
Nara Womens Univ. |
A convex formulations of learning from crowds Hiroshi Kajino, Hisashi Kashima (UT) IBISML2011-76 |
It has attracted considerable attention to use crowdsourcing services
to collect a large amount of labeled data for ma... [more] |
IBISML2011-76 pp.231-236 |
IBISML |
2011-11-10 15:45 |
Nara |
Nara Womens Univ. |
A Linear Time Subpath Kernel for Trees Daisuke Kimura, Hisashi Kashima (Univ. of Tokyo) IBISML2011-85 |
Kernel method is one of the promising approaches to learning with
tree-structured data, and various efficient tree ker... [more] |
IBISML2011-85 pp.291-296 |
IBISML |
2011-06-20 10:35 |
Tokyo |
Takeda Hall |
On the Convergence of Convex Tensor Estimation Ryota Tomioka, Taiji Suzuki (Univ. Tokyo), Kohei Hayashi (NAIST), Hisashi Kashima (Univ. Tokyo) IBISML2011-14 |
凸最適化に基づくテンソル分解アルゴリズムの統計的な性能について解析し,報
告する.従来テンソル分解は非凸の最適化問題として定式化され,そのため性
能の解析は困難であった.本論文では,ある条件のもとで,推定されたテンソ
ルを$\h... [more] |
IBISML2011-14 pp.97-102 |
IBISML |
2011-03-29 10:10 |
Osaka |
Nakanoshima Center, Osaka Univ. |
Tensor factorization using auxiliary information Atsuhiro Narita (Univ. Tokyo), Kohei Hayashi (NAIST), Ryota Tomioka (Univ. Tokyo), Hisashi Kashima (Univ. of Tokyo/JST) IBISML2010-124 |
Most of the existing completion methods of tensors (i.e. multi-way
arrays) only assume that tensors to be completed are... [more] |
IBISML2010-124 pp.139-146 |