Committee |
Date Time |
Place |
Paper Title / Authors |
Abstract |
Paper # |
IBISML, NC, IPSJ-BIO, IPSJ-MPS [detail] |
2024-06-21 10:55 |
Okinawa |
OIST (Okinawa) |
Analysis of the model selection method sBIC with learning coefficients and its applications Keita Yamazaki, Tomoyasu Ohba, Haru Kobayashi, Kyousuke Shimizu (Nihon Univ.), Daisuke Kaji (DENSO), Miki Aoyagi (Nihon Univ.) NC2024-16 IBISML2024-16 |
In recent years, it has been demonstrated that the asymptotic behaviors of generalization error, empirical error, and fr... [more] |
NC2024-16 IBISML2024-16 pp.98-104 |
PRMU |
2022-12-16 16:15 |
Toyama |
Toyama International Conference Center (Toyama, Online) (Primary: On-site, Secondary: Online) |
On partitioning of the space of one-dimensional normal distributions based on e(m)-divergence Jun Fujiki (Fukuoka Univ.), Shotaro Akaho (AIST) PRMU2022-54 |
Partitioning of the space of one-dimensional normal distributions based on e(m)-divergence is useful when clustering and... [more] |
PRMU2022-54 pp.116-121 |
AP, SANE, SAT (Joint) |
2022-07-29 11:15 |
Hokkaido |
Asahikawa Taisetsu Crystal Hall (Hokkaido, Online) (Primary: On-site, Secondary: Online) |
Evaluation of Secure Transmission Performance of Spatially-Selective Modulation in an Indoor Environment Hitoro Yonawa, Yuka Morishima, Hisato Iwai, Shinsuke Ibi (Doshisha Univ.) AP2022-57 |
As a physical layer security technique, we have proposed spatially-selective modulation (SSM) which consists of multiple... [more] |
AP2022-57 pp.129-134 |
NLP, MICT, MBE, NC (Joint) [detail] |
2022-01-23 11:20 |
Online |
Online (Online) |
Towards 2-class classification via e(m)-PCA Jun Fujiki (Fukuoka Univ.), Shotaro Akaho (AIST) NC2021-43 |
In many data analysis methods, observed data are treated as points in high-dimensional Euclidean space, called feature s... [more] |
NC2021-43 pp.55-58 |
IBISML |
2022-01-17 11:00 |
Online |
Online (Online) |
Cluster approximation in quantum Boltzmann machine based on information geometry Masaya Hoshikawa, Tomohiro Ogawa (UEC) IBISML2021-21 |
A Boltzmann Machine (BM) is a model of machine learning which consists
of mutually connected probabilistic binary units... [more] |
IBISML2021-21 pp.23-28 |
QIT (2nd) |
2021-11-30 13:30 |
Online |
Online (Online) |
[Poster Presentation]
Non-monotone metric on the quantum parametric model Jun Suzuki (UEC) |
In this paper, we study a family of quantum Fisher metrics based on a convex mixture of two well-known inner products, w... [more] |
|
NC, IBISML, IPSJ-BIO, IPSJ-MPS [detail] |
2021-06-28 16:50 |
Online |
Online (Online) |
Constellation of time dependent data using information geometry Taiki Sugiura, Takuma Fuse (Recruit), Noboru Murata (Waseda Univ.) NC2021-9 IBISML2021-9 |
Constellation is an essential technique for gaining knowledge from large amounts of high-dimensional data and understand... [more] |
NC2021-9 IBISML2021-9 pp.62-69 |
IE, ITS, ITE-MMS, ITE-ME, ITE-AIT [detail] |
2021-02-19 13:45 |
Online |
Online (Online) |
[Special Talk]
The data science educator and mathematics
-- from graduate student in mathematics viewpoint -- Naomichi Nakajima (Hokkaido Univ.) |
At the Education and Research Center for Mathematical and Data Science, Hokkaido University (MDS center), education of t... [more] |
|
IBISML |
2020-01-09 13:50 |
Tokyo |
ISM (Tokyo) |
Dimensionality reduction method for gaussian process posteriors based on information geometry Hideaki Ishibashi (Kyutech), Shotaro Akaho (AIST/RIKEN) IBISML2019-20 |
This paper proposes an extension of principal component analysis for gaussian process posteriors which is denoted by GP-... [more] |
IBISML2019-20 pp.17-24 |
SIP, EA, SP, MI (Joint) [detail] |
2018-03-19 13:40 |
Okinawa |
(Okinawa) |
[Short Paper]
Proposal of a spatio-temporal statistical model of generated anatomical landmarks Aoi Shinjo, Atsushi Saito (TUAT), Tetsuya Takakuwa, Shigehito Yamada (Kyoto Univ.), Hiroshi Matsuzoe, Hidekata Hontani (NITech), Akinobu Shimizu (TUAT) MI2017-77 |
(To be available after the conference date) [more] |
MI2017-77 pp.41-42 |
WBS, IT, ISEC |
2018-03-08 09:25 |
Tokyo |
Katsusika Campas, Tokyo University of Science (Tokyo) |
Information Geometrical Study of Cluster-Model Approximation for Boltzmann Machines Kenta Toyoda, Tomohiro Ogawa (Univ. of Electro- Commu.) IT2017-104 ISEC2017-92 WBS2017-85 |
In learning algorithms for Boltzmann machines, it is necessary but very hard to calculate the expectation of units with ... [more] |
IT2017-104 ISEC2017-92 WBS2017-85 pp.7-12 |
IBISML |
2018-03-05 14:15 |
Fukuoka |
Nishijin Plaza, Kyushu University (Fukuoka) |
Exponential Family of Markov Kernels and Asymptotic Exponential Family of Markov Sources Jun'ichi Takeuchi (Kyushu Univ.), Hiroshi Nagaoka (UEC) IBISML2017-93 |
For parametric models of Markov sources, we prove that the notion of asymptotic exponential family is equivalent to the ... [more] |
IBISML2017-93 pp.21-25 |
MBE, NC (Joint) |
2017-11-24 16:50 |
Miyagi |
Tohoku University (Miyagi) |
NC2017-32 |
Continuous latent variable model is a category of dimension reduction methods, which estimates low dimensional latent va... [more] |
NC2017-32 pp.29-34 |
IT |
2017-07-14 15:10 |
Chiba |
Chiba University (Chiba) |
On Information Geometry of Tree Models Jun'ichi Takeuchi (Kyushu Univ.), Hiroshi Nagaoka (UEC) IT2017-35 |
We discuss information geometrical properties of the familes of Markov sources defined based on context trees. [more] |
IT2017-35 pp.109-113 |
QIT (2nd) |
2017-05-30 10:10 |
Kyoto |
Ritesumeikan University, Suzaku Campus (Kyoto) |
Total Detection Probability in Generalized Probability Theories
-- channel capacity and geometry of the state space -- Keiji Matsumoto (STI), Gen Kimura (SIT) |
The qubit state space is a beautiful sphere, but the high-dimensional one is increasingly asymmetric in proportion to th... [more] |
|
QIT (2nd) |
2016-11-24 13:00 |
Ibaraki |
KEK Kobayashi-hall (Ibaraki) |
[Poster Presentation]
Data processing for qubit state tomography: An information geometric approach Koichi Yamagata (Chuo Univ), Akio Fujiwara (Osaka Univ) |
A statistically feasible data post-processing method
for the conventional qubit state tomography is studied from an in... [more] |
|
IBISML |
2016-11-17 17:00 |
Kyoto |
Kyoto Univ. (Kyoto) |
[Poster Presentation]
How to make Tsallis entropy become an additive entropy Masaru Tanaka (Fukuoka Univ.) IBISML2016-99 |
$tau$-information geometry, that is a new formulation of information geometry, is given by extending a translation opera... [more] |
IBISML2016-99 pp.381-386 |
COMP |
2016-03-14 14:20 |
Fukuoka |
(Fukuoka) |
[Fellow Memorial Lecture]
Computational Geometry and its Applications to Geographic Information System Keiko Imai (Chuo Univ.) COMP2015-39 |
Geographic Information System (GIS) is one of the application areas of Computational Geometry. Maps are the most basic ... [more] |
COMP2015-39 p.21 |
IBISML |
2015-11-26 15:00 |
Ibaraki |
Epochal Tsukuba (Ibaraki) |
[Poster Presentation]
Non-parametric e-mixture Estimation Ken Takano (Waseda Univ.), Hideitsu Hino (Univ. of Tsukuba), Shotaro Akaho (AIST), Noboru Murata (Waseda Univ.) IBISML2015-59 |
Mixture modeling is one of the simplest ways to represent complicated probability density functions, and to integrate in... [more] |
IBISML2015-59 pp.45-52 |
IBISML |
2015-11-26 15:00 |
Ibaraki |
Epochal Tsukuba (Ibaraki) |
[Poster Presentation]
On a family of $q$-normal distributions in the framework of $tau$-information geometry Masaru Tanaka (Fukuoka Univ.) IBISML2015-62 |
$tau$-information geometry, that is a new formulation of information geometry, is given by extending a translation opera... [more] |
IBISML2015-62 pp.69-73 |