Committee |
Date Time |
Place |
Paper Title / Authors |
Abstract |
Paper # |
IBISML |
2017-11-09 13:00 |
Tokyo |
Univ. of Tokyo |
Robust one dimensional phase unwrapping using Markov random fields Yasuhisa Nakashima (Univ. Tokyo), Yasuhiko Igarashi (JST), Yasushi Naruse (NICT), Masato Okada (Univ. Tokyo) IBISML2017-45 |
In the measurement of crustal deformation using satellite or aircraft sensors, interferometric synthetic aperture radar ... [more] |
IBISML2017-45 pp.77-84 |
IBISML |
2017-11-10 13:00 |
Tokyo |
Univ. of Tokyo |
[Poster Presentation]
Bin width optimization of multidimensional histogram on count data Kensuke Muto, Hirotaka Sakamoto, Keisuke Matsuura, Takahisa Arima, Masato Okada (Tokyo Univ.) IBISML2017-69 |
A large amount of 4-dimensional count data are obtained by inelastic neutron scattering experiments conducted by chopper... [more] |
IBISML2017-69 pp.255-260 |
IBISML |
2017-11-10 13:00 |
Tokyo |
Univ. of Tokyo |
[Poster Presentation]
Nonlinear parametric model based on power law for tsunami height prediction Masashi Yoshikawa (UT), Yasuhiko Igarashi (JST), Shin Murata (UT), Toshitaka Baba (TU), Takane Hori (JAMSTEC), Masato Okada (UT) IBISML2017-70 |
When a subduction-zone earthquake occurs, we need to predict the tsunami height in order to cope with the tsunami damage... [more] |
IBISML2017-70 pp.261-267 |
IBISML |
2017-11-10 13:00 |
Tokyo |
Univ. of Tokyo |
[Poster Presentation]
Proposal of λ-scan Method in Spectral Deconvolution Yohachi Mototake (Univ of Tokyo), Yasuhiko Igarashi (NIMS), Hikaru Takenaka (Univ of Tokyo), Kenji Nagata (AIST), Masato Okada (Univ of Tokyo) IBISML2017-80 |
Spectral deconvolution is a method to fit spectral data as the sum of unimodal basis functions and is a useful method in... [more] |
IBISML2017-80 pp.325-332 |
IBISML |
2017-11-10 13:00 |
Tokyo |
Univ. of Tokyo |
Approximated hyperparameter distribution estimation using Gaussian process and Bayesian optimization Shun Katakami, Hirotaka Sakamoto, Masato Okada (UTokyo) IBISML2017-81 |
In order to reduce the computational cost of Bayesian inference, we propose a method to estimate the Bayesian posterior ... [more] |
IBISML2017-81 pp.333-338 |
IBISML |
2017-11-10 13:00 |
Tokyo |
Univ. of Tokyo |
Statistical Mechanical Analysis of Learning with Two-Layer Perceptron with Multiple Output Units
-- Reconsidering Plateau Phenomenon -- Yuki Yoshida (UTokyo), Ryo Karakida (AIST), Masato Okada (UTokyo/AIST/RIKEN BSI), Shun-ichi Amari (RIKEN BSI) IBISML2017-83 |
The plateau phenomenon --- stopping decrease of error in the middle of learning --- is problematic.
Since Amari et al. ... [more] |
IBISML2017-83 pp.347-354 |
CCS |
2017-03-11 09:55 |
Tokyo |
ELSI, TITECH |
One dimensional phase inference by sequential Bayesian filter Hiroaki Umehara (NICT), Masato Okada (UTokyo), Yasushi Naruse (NICT) CCS2016-50 |
[more] |
CCS2016-50 pp.31-36 |
NC, NLP (Joint) |
2017-01-26 15:00 |
Fukuoka |
Kitakyushu Foundation for the Advanement of Ind. Sci. and Tech. |
Statistical mechanics of coherent Ising machine
-- The case of ferromagnetic and finite loading Hopfield models -- Toru Aonishi (Tokyo Tech.), Kazushi Mimura (Hiroshima City Univ.), Shoko Utsunomiya (NII), Masato Okada (Univ. of Tokyo), Yoshihisa Yamamoto (NII/Stanford Univ.) NC2016-50 |
The coherent Ising machine (CIM) is attracted attention as one of most effective Ising computing architecture for solvin... [more] |
NC2016-50 pp.13-18 |
NC, NLP (Joint) |
2017-01-26 16:00 |
Fukuoka |
Kitakyushu Foundation for the Advanement of Ind. Sci. and Tech. |
Fast Receptive field Inference with Sparse Fourirer Representation by using LASSO Takeshi Tanida, Hirotaka Sakamoto, Yasuhiko Igarashi, Takeshi Ideriha, Satoru Tokuda (Univ. of Tokyo), Kota Sasaki, Izumi Ohzawa (Osaka Univ.), Masato Okada (Univ. of Tokyo/RIKEN) NC2016-52 |
We propose fast receptive eld(RF) inference. The RF describes how a neuron sums up its inputs across
space and time. T... [more] |
NC2016-52 pp.25-30 |
IBISML |
2016-11-16 15:00 |
Kyoto |
Kyoto Univ. |
Statistical Mechanical Analysis of Fast Online Learning with Weight Normalization Yuki Yoshida, Ryo Karakida, Masato Okada (UTokyo), Shun-ichi Amari (RIKEN) IBISML2016-60 |
Weight normalization (WN), a newly developed optimization algorithm for neural networks by Salimans & Kingma(2016), fact... [more] |
IBISML2016-60 pp.101-108 |
IBISML |
2016-11-16 15:00 |
Kyoto |
Kyoto Univ. |
Inference of Classical Spin Model by Multidimensional Multiple Histogram Method Hikaru Takenaka (UTokyo), Kenji Nagata (UTokyo/AIST/JST), Takashi Mizokawa (Waseda Univ.), Masato Okada (UTokyo/RIKEN) IBISML2016-61 |
We propose a novel method for effective Bayesian inference of classical spin model by the multidimensional multiple hist... [more] |
IBISML2016-61 pp.109-116 |
IBISML |
2016-11-17 14:00 |
Kyoto |
Kyoto Univ. |
Gaussian Markov random field model without periodic boundary conditions Shun Katakami, Hirotaka Sakamoto, Shin Murata, Masato Okada (UTokyo) IBISML2016-83 |
In this study, we discuss Gaussian Markov random field model without periodic boundary conditions. First, we formulate a... [more] |
IBISML2016-83 pp.267-274 |
IBISML |
2016-11-17 14:00 |
Kyoto |
Kyoto Univ. |
Extraction of low dimensional attractors embedded in a recurrent neural network by using Dynamic Mode Decomposition Shin Murata (Univ. Tokyo), Masato Okada (Univ. Tokyo/RIKEN) IBISML2016-85 |
Dynamic Mode Decomposition (DMD) decomposes high-dimensional dynamical data into a few dynamic modes and has been develo... [more] |
IBISML2016-85 pp.279-285 |
IBISML |
2016-11-17 14:00 |
Kyoto |
Kyoto Univ. |
Exhaustive search for sparse variable selection in linear regression Yasuhiko Igarashi, Hikaru Takenaka, Nakanishi-Ohno Yoshinori (UTokyo), Makoto Uemura (Hiroshima Univ.), Shiro Ikeda (ISM), Masato Okada (UTokyo) IBISML2016-90 |
We proposed the $K$-sparse
Exhaustive-Search (ES-$K$) method,
in which, assuming the optimum combination of
explan... [more] |
IBISML2016-90 pp.313-320 |
IBISML |
2016-11-17 14:00 |
Kyoto |
Kyoto Univ. |
An Exhaustive Search with Support Vector Machine (ES-SVM) for sparse variable selection Daiki Kawabata (UTokyo), Hiroko Ichikawa (TUS), Yasuhiko Igarashi (UTokyo), Kenji Nagata (AIST/JST/UTokyo), Satoshi Eifuku, Ryoi Tamura (Toyama Univ.), Masato Okada (UTokyo) IBISML2016-96 |
Nagata et al.(2015) has proposed Exhaustive Search with Support Vector Machine(ES-SVM) which calculates a cross validati... [more] |
IBISML2016-96 pp.361-368 |
NC, IPSJ-BIO, IBISML, IPSJ-MPS [detail] |
2016-07-05 16:25 |
Okinawa |
Okinawa Institute of Science and Technology |
Sparse Simultaneous Estimation of Phase Response Curves using Spike-Triggered Average Man Arakaki, Yasuhiko Igarashi (Univ. Tokyo), Toshiaki Omori (Kobe Univ.), Masato Okada (Univ. Tokyo/RIKEN) NC2016-7 |
Phase response curves (PRC) have been used extensively to estimate the response properties of neurons.
A recent measure... [more] |
NC2016-7 pp.165-170 |
MBE, NC (Joint) |
2016-03-22 17:00 |
Tokyo |
Tamagawa University |
[Invited Talk]
From associative memory model to sparse modeling and brain-like artificial intelligence Masato Okada (UTokyo) NC2015-79 |
In this talk, I first introduce statistical-mechanical informatics. In my opinion, statistical-mechanical informatics ha... [more] |
NC2015-79 p.57 |
MBE, NC (Joint) |
2016-03-23 14:00 |
Tokyo |
Tamagawa University |
External Input Response of Neural Network Model with Irregular Spontaneous Activity Yoshihiro Nagano, Ryo Karakida (UTokyo), Norifumi Watanabe (TUT), Atsushi Aoyama (Keio Univ), Masato Okada (UTokyo) NC2015-85 |
The previous studies have reported that the neural network model with connections following a lognormal distribution rep... [more] |
NC2015-85 pp.89-94 |
IBISML |
2015-11-27 14:00 |
Ibaraki |
Epochal Tsukuba |
[Poster Presentation]
Minimum required data amount in Bayesian inference from the viewpoint of specific heat Satoru Tokuda, Kenji Nagata, Masato Okada (Univ. of Tokyo) IBISML2015-74 |
The accuracy of Bayesian inference depends on the number of samples or noise. Sample size or noise level often changes t... [more] |
IBISML2015-74 pp.159-166 |
CCS |
2015-11-09 14:25 |
Kyoto |
Inamori Foundation Memorial Building, Kyoto Univ. |
A macroscopic neural mass model constructed from a current-based network of spiking neurons Hiroaki Umehara (NICT), Masato Okada (UTokyo), Jun-nosuke Teramae (Osaka Univ.), Yasushi Naruse (NICT) CCS2015-51 |
Neural mass model describing the dynamics of mean membrane potentials for neuronal populations are formulated from the n... [more] |
CCS2015-51 pp.35-40 |