| IEICE Technical Committee Submission System Conference Schedule |
Online Proceedings [Sign in] Tech. Rep. Archives |
All Technical Committee Conferences (Searched in: All Years)
| Committee | Date Time | Place | Paper Title / Authors | Abstract | Paper # | |
| IBISML | 2024-12-20 10:30 |
Hokkaido | Lecture room 1 (D101), Graduate School of Environmental Science (Hokkaido, Online) (Primary: On-site, Secondary: Online) |
Training Deep Neural Networks for Fast Data Assimilation Yuta Ono (UTokyo), Yuta Tarumi, Keisuke Fukuda, Shin-ichi Maeda (PFN)  IBISML2024-36 |
Data assimilation (DA) is imperative to achieve accurate state estimation from observations and numerical simulations. A... [more] | IBISML2024-36 pp.34-40 |
| SeMI, RCS, NS, SR, RCC (Joint) |
2019-07-10 16:45 |
Osaka | I-Site Nanba（Osaka） (Osaka) | [Invited Talk]
Current Status of Reinforcement Learning
-- Algorithms and Applications -- Shin-ichi Maeda (PFN) RCC2019-18 NS2019-51 RCS2019-108 SR2019-27 SeMI2019-27 |
Reinforcement Learning is a framework to optimize an action sequence in terms of the return maximization. In this talk, ... [more] | RCC2019-18 NS2019-51 RCS2019-108 SR2019-27 SeMI2019-27 p.39(RCC), p.49(NS), p.43(RCS), p.49(SR), p.53(SeMI) |
| IBISML | 2016-03-17 16:10 |
Tokyo | Institute of Statistical Mathematics (Tokyo) | Bayesian Monte-Carlo tree search method and its application to linear constrained nonlinear control problems Ryo Otsuki, Shin-ichi Maeda, Shin Ishii (Kyoto Univ.) IBISML2015-98 |
For the nonlinear dynamics where the state transition is nonlinear with respect to the input, in general, we cannot obta... [more] | IBISML2015-98 pp.31-38 |
| IBISML | 2015-11-27 14:00 |
Ibaraki | Epochal Tsukuba (Ibaraki) | [Poster Presentation]
Regularization by local distributional smoothing Takeru Miyato, Shin-ichi Maeda, Masanori Koyama, Ken Nakae, Shin Ishii (Kyoto Univ.) IBISML2015-87 |
Smoothness regularization is a popular method to decrease generalization error. We propose a novel regularization techni... [more] | IBISML2015-87 pp.257-264 |
| IBISML | 2015-11-27 14:00 |
Ibaraki | Epochal Tsukuba (Ibaraki) | [Poster Presentation]
Bayesian Masking for Sparse Feature Selection Yohei Kondo (Kyoto Univ.), Kohei Hayashi (NII), Shin-ichi Maeda (Kyoto Univ.) IBISML2015-88 |
In linear regression, we can reduce the weights of irrelevant features by L2 or L1 regularization. However, such a regul... [more] | IBISML2015-88 pp.265-272 |
| NC, MBE (Joint) |
2014-03-18 14:00 |
Tokyo | Tamagawa University (Tokyo) | Fusion of Multiple Cues from Color and Depth Domains using Occlusion Aware Bayesian Tracker Kourosh Meshgi, Shin-ichi Maeda, Shigeyuki Oba, Shin Ishii (Kyoto Univ.) NC2013-110 |
[more] | NC2013-110 pp.127-132 |
| NC, MBE (Joint) |
2014-03-18 14:20 |
Tokyo | Tamagawa University (Tokyo) | 3D Superresolution of Microscopic Images based on Variational Bayesian Inference via Chebyshev polynomials approximation Yasuhiro Imoto, Shin-ichi Maeda, Shin Ishii (Kyoto Univ.) NC2013-111 |
Optical microscopes are used to elucidate changes in cellular functions mediated by morphological changes of cells in vi... [more] | NC2013-111 pp.133-138 |
| NC, MBE (Joint) |
2014-03-18 13:00 |
Tokyo | Tamagawa University (Tokyo) | Layered Monte-Carlo planning method for incomplete information games and its application to Puyo-Puyo Ryo Otsuki, Shin-ichi Maeda, Shin Ishii (Kyoto Univ.) NC2013-136 |
Recently, the effectiveness of the game-tree search algorithm based on UCT and its applicability to the game Go have bee... [more] | NC2013-136 pp.275-280 |
| NC, MBE (Joint) |
2014-03-18 15:00 |
Tokyo | Tamagawa University (Tokyo) | Human Activity Rcognition with Skeleton-data and Structured Hierarchical Hidden Markov Model Tomoji Sawada, Shin-ichi Maeda, Shin Ishii (Kyoto Univ.) NC2013-141 |
Recognition of human activities nowadays is used not only for monitoring system, but also interface, motion analysis of... [more] |
NC2013-141 pp.305-310 |
| PRMU, IBISML, IPSJ-CVIM [detail] | 2013-09-02 17:30 |
Tottori | (Tottori) | Enhancing Probabilistic Appearance-Based Object Tracking with Depth Information
-- Object Tracking under Occlusion -- Kourosh Meshgi, Yu-zhe Li, Shigeyuki Oba, Shin-ichi Maeda, Shin Ishii (Kyoto Univ.) PRMU2013-42 IBISML2013-22 |
Object tracking has attracted recent attention because of high demands for its everyday-life applications. Handling occl... [more] | PRMU2013-42 IBISML2013-22 pp.85-91 |
| NC, NLP | 2013-01-24 10:30 |
Hokkaido | Hokkaido University Centennial Memory Hall (Hokkaido) | Control of the falling cat motion by using path-integral reinforcement learning Daichi Nakano, Shin-ichi Maeda, Shin Ishii (Kyoto Univ) NLP2012-107 NC2012-97 |
The falling-cat motion is a motion for controlling the cat's posture under no existence of external force. To obtain a c... [more] | NLP2012-107 NC2012-97 pp.19-24 |
| NC, NLP | 2013-01-24 10:50 |
Hokkaido | Hokkaido University Centennial Memory Hall (Hokkaido) | Significance of non-stationary of dynamics for learning cooperative behavior Akihiro Tawa, Shin-ichi Maeda, Shin Ishii (Kyoto Univ.) NLP2012-108 NC2012-98 |
To understand how cooperative behaviors emerge is important in the field of multi-agent system research. Although this e... [more] | NLP2012-108 NC2012-98 pp.25-30 |
| NC, NLP | 2013-01-24 11:10 |
Hokkaido | Hokkaido University Centennial Memory Hall (Hokkaido) | depth estimation from microscopic images using Bayesian inference Yasuhiro Imoto, Shin-ichi Maeda, Shin Ishii (Kyoto Univ.) NLP2012-109 NC2012-99 |
In cellular biology, it is important to know 3D cellular shape to understand the cellular function. However, existing mi... [more] | NLP2012-109 NC2012-99 pp.31-36 |
| MI | 2010-11-15 13:25 |
Kyoto | Shimadzu Corp. (Kyoto) | Bayesian modeling of medical X-ray computed tomography Shin-ichi Maeda (Kyoto Univ.), Atsunori Kanemura (ATR), Shin Ishii (Kyoto Univ.) MI2010-73 |
The tradeoff between the resolution of CT images and the amount of exposure to radiation leads us to desire a CT algorit... [more] | MI2010-73 pp.39-44 |
| IBISML | 2010-06-15 15:10 |
Tokyo | Takeda Hall, Univ. Tokyo (Tokyo) | Generalization of TD-learning from a Semiparametric Statistical Viewpoint Tsuyoshi Ueno, Shin-ichi Maeda (Kyoto Univ.), Motoaki Kawanabe (Fraunhofer First), Shin Ishii (Kyoto Univ.) IBISML2010-20 |
(Advance abstract in Japanese is available) [more] | IBISML2010-20 pp.131-138 |
| NC, MBE (Joint) |
2010-03-10 11:05 |
Tokyo | Tamagawa University (Tokyo) | Bayesian X-ray Computed Tomography Using a Mixture Prior Wataru Fukuda, Shin-ichi Maeda, Atsunori Kanemura, Shin Ishii (Kyoto Univ.) NC2009-133 |
[more] | NC2009-133 pp.267-272 |
| NC, MBE (Joint) |
2010-03-11 11:05 |
Tokyo | Tamagawa University (Tokyo) | Learning of Go board state evaluation by online adaptive natural gradient method Hiroki Tomizawa, Shin-ichi Maeda, Shin Ishii (Kyoto Univ.) NC2009-164 |
We propose a supervised learning of Go board state evaluation function with many game records of experts'. By training a... [more] | NC2009-164 pp.449-454 |
| SIP, CAS, CS | 2010-03-02 13:45 |
Okinawa | Hotel Breeze Bay Marina, Miyakojima (Okinawa) | [Poster Presentation]
Geometrical analysis of linear discriminant analysis algorithms and instrument feature extraction Mizuki Ihara, Kazushi Ikeda (NAIST), Shin-ichi Maeda (Kyoto Univ.) CAS2009-123 SIP2009-168 CS2009-118 |
Extracting only the essential sound attributes from sounds is one of the fundamental issues of music information retriev... [more] | CAS2009-123 SIP2009-168 CS2009-118 pp.243-244 |
| NC, MBE (Joint) |
2009-03-11 16:35 |
Tokyo | Tamagawa Univ. (Tokyo) | A Learning Algorithm of Helmholtz Machine with Mean Field Approximation Yuki Aoki (Nara Inst. of Sci and Tech.), Shin-ichi Maeda, Shin Ishii (Kyoto Univ.) NC2008-113 |
It is often required to extract a compact feature of an original high-dimensional datum. Such a compactfeature is useful... [more] | NC2008-113 pp.57-62 |
| NC, MBE (Joint) |
2009-03-12 15:15 |
Tokyo | Tamagawa Univ. (Tokyo) | Semiparametric Statistical Approach to Value Function Estimation Tsuyoshi Ueno (Kyoto Univ.), Motoaki Kawanabe (Fraunhofer First), Takeshi Mori, Shin-ichi Maeda, Shin Ishii (Kyoto Univ.) NC2008-146 |
Recently least-squares temporal difference (LSTD) learning has been developed for the model-free value function es... [more] |
NC2008-146 pp.255-260 |
[Return to Top Page]