Committee |
Date Time |
Place |
Paper Title / Authors |
Abstract |
Paper # |
NC, IBISML, IPSJ-BIO, IPSJ-MPS [detail] |
2023-06-30 11:10 |
Okinawa |
OIST Conference Center (Primary: On-site, Secondary: Online) |
On performance degradation of a method by minimizing the conditional mutual information for the out-of-distribution generalization Genki Takahashi, Toshiyuki Tanaka (Kyoto University) NC2023-15 IBISML2023-15 |
In the out-of-distribution generalization problem, the smaller the degree of change in the data generating distribution ... [more] |
NC2023-15 IBISML2023-15 pp.91-97 |
PRMU, IBISML, IPSJ-CVIM [detail] |
2023-03-02 10:55 |
Hokkaido |
Future University Hakodate (Primary: On-site, Secondary: Online) |
Non-uniqueness of solutions in Physics-informed neural networks Shuichiro Tsuda, Toshiyuki Tanaka (Kyoto Univ.) PRMU2022-81 IBISML2022-88 |
PINN is a method of acquiring a neural network that represents the solution of a differential equation by including a co... [more] |
PRMU2022-81 IBISML2022-88 pp.125-128 |
SIP, IT, RCS |
2021-01-22 14:50 |
Online |
Online |
Performance Analysis of Matrix Sketching for Principal Component Analysis Fumito Tagashira, Tomoyuki Obuchi, Toshiyuki Tanaka (Kyoto Univ.) IT2020-107 SIP2020-85 RCS2020-198 |
[more] |
IT2020-107 SIP2020-85 RCS2020-198 pp.247-252 |
IBISML |
2020-03-11 15:35 |
Kyoto |
Kyoto University (Cancelled but technical report was issued) |
Regret analysis of Thompson sampling using a general beta prior Yuto Kawamura, Toshiyuki Tanaka (Kyoto Univ.) IBISML2019-49 |
For Bernoulli bandits, the asymptotic optimality of Thompson sampling with the uniform prior in terms of the regret has ... [more] |
IBISML2019-49 pp.107-112 |
IBISML |
2019-03-06 11:30 |
Tokyo |
RIKEN AIP |
Optimal Kernel for Mode Estimation via Kernel Density Estimation Ryoya Yamasaki, Toshiyuki Tanaka (Kyoto Univ.) IBISML2018-113 |
We have derived kernel functions that minimize the asymptotic mean squared error of the mode estimate, which is defined ... [more] |
IBISML2018-113 pp.59-64 |
IBISML |
2018-11-05 15:10 |
Hokkaido |
Hokkaido Citizens Activites Center (Kaderu 2.7) |
[Poster Presentation]
How does the complexity of critic affect the performance of WGAN? Akihiro Iohara, Toshiyuki Tanaka (Kyoto Univ.) IBISML2018-60 |
WGAN is a generative model that learns by minimizing the Wasserstein distance between the generator distribution and the... [more] |
IBISML2018-60 pp.119-125 |
IBISML |
2018-11-05 15:10 |
Hokkaido |
Hokkaido Citizens Activites Center (Kaderu 2.7) |
[Poster Presentation]
Regularized Modal Regression via Iteratively Reweighted Least-Squares Ryoya Yamasaki, Toshiyuki Tanaka (Kyoto Univ.) IBISML2018-88 |
In modal regression, iterative methods that are based on the modal EM (MEM) and those based on the iteratively reweighte... [more] |
IBISML2018-88 pp.329-335 |
IBISML |
2017-11-09 13:00 |
Tokyo |
Univ. of Tokyo |
Efficient modal regression using gradient ascent and descent method Ryoya Yamasaki, Toshiyuki Tanaka (Kyoto Univ.) IBISML2017-57 |
Nonparametric modal regression is a method for regression analysis, which can estimate local maxima of the conditional p... [more] |
IBISML2017-57 pp.169-176 |
IBISML |
2016-11-16 15:00 |
Kyoto |
Kyoto Univ. |
Efficient learning of ranking model using belief propagation Arise Kuriya, Toshiyuki Tanaka (Kyoto Univ.) IBISML2016-64 |
In the area of Learning to Rank, the models whose output is ranking are trained from data.
The exponential model propos... [more] |
IBISML2016-64 pp.131-135 |
IBISML |
2016-11-16 15:00 |
Kyoto |
Kyoto Univ. |
Machine learning approach to glottal closure instant estimation from speech signals Akihiro Iohara, Toshiyuki Tanaka (Kyoto Univ.) |
[more] |
|
NS, IN (Joint) |
2016-03-03 10:40 |
Miyazaki |
Phoenix Seagaia Resort |
Throughput Equalization for CSMA-based Wireless Networks using Mean-Field Hard-Core Model Yiyan Liu, Toshiyuki Tanaka (Kyoto Univ.), K. Shashi Prabh (Shiv Nadar Univ.) IN2015-116 |
In this paper, we consider the problem of throughput equalization for wireless devices in CSMA-based wireless networks. ... [more] |
IN2015-116 pp.49-54 |
IBISML |
2015-11-27 14:00 |
Ibaraki |
Epochal Tsukuba |
[Poster Presentation]
Performance degradation of AMP for Ising perceptron when the system size is small Arise Kuriya, Toshiyuki Tanaka (Kyoto Univ.) IBISML2015-85 |
Approximate Massage Passing (AMP) algorithm, proposed by Donoho et al., is derived from Belief Propagation (BP) algorith... [more] |
IBISML2015-85 pp.241-247 |
IBISML |
2015-03-06 11:00 |
Kyoto |
Kyoto University |
Asymptotic Properties of Area Under the ROC Curve via Likelihood Ratio Based Ranking Function Kentaro Nakanishi, Toshiyuki Tanaka (Kyoto Univ.), Naonori Ueda (NTT) IBISML2014-92 |
We study properties of the area under the ROC curve, which is one of measures to evaluate performance of a binary classi... [more] |
IBISML2014-92 pp.55-62 |
MI |
2015-03-02 13:40 |
Okinawa |
Hotel Miyahira |
Visualization of Lenticulostriate Artery by MR Angiography
-- Sparse Sampling Rates and Compressed Sensing Reconstruction -- Tomohisa Okada, Koji Fujimoto, Akira Yamamoto, Yasutaka Fushimi, Kei Sano, Toshiyuki Tanaka, Kaori Togashi (Kyoto Univ.) MI2014-69 |
Deep perforating arteries including lenticulostriate atery are causative of 1/3 of all strokes. However, these arteirs h... [more] |
MI2014-69 pp.83-84 |
MI |
2015-03-03 09:48 |
Okinawa |
Hotel Miyahira |
Quality Improvement of SPECT Images via Random Coded Aperture and Compressed Sensing Takenori Segawa, Toshiyuki Tanaka (Kyoto Univ.) MI2014-92 |
There is a trade-off between quality of reconstructed images and
discomfort of patients, caused by long scan time and... [more] |
MI2014-92 pp.183-188 |
MI |
2015-03-03 10:10 |
Okinawa |
Hotel Miyahira |
TOF-MRA image reconstruction by compressed sensing technique
-- comparison of three different regularization techniques -- Akira Yamamoto, Koji Fujimoto, Yasutaka Fushimi, Tomohisa Okada, Kei Sano, Toshiyuki Tanaka, Kaori Togashi (Kyoto Univ.) MI2014-93 |
Three different regularization terms, L1-norm of signal intensity, L1-norm of wavelet coefficients, and total variation,... [more] |
MI2014-93 pp.189-192 |
SIP, EA, SP |
2015-03-03 14:40 |
Okinawa |
|
Acoustic feature analysis of desirable voice in chorus Akihiro Tawa, Toshiyuki Tanaka (Kyoto Univ.) EA2014-126 SIP2014-167 SP2014-189 |
In this study, we analyzed characteristics of the ``bright voice'' which is desirable in chorus.
Voice timbre correspon... [more] |
EA2014-126 SIP2014-167 SP2014-189 pp.301-306 |
IT |
2014-12-09 14:50 |
Toyama |
Unazuki New Otani Hotel |
[Invited Talk]
Compressed Sensing
-- A paradigm of information processing based on sparsity -- Toshiyuki Tanaka (Kyoto Univ.) IT2014-51 |
[more] |
IT2014-51 pp.7-12 |
IBISML |
2014-11-17 17:00 |
Aichi |
Nagoya Univ. |
[Poster Presentation]
Application of partial parallel interference cancellation to sparse signal recovery Koujin Takeda (Ibaraki Univ.), Toshiyuki Tanaka (Kyoto Univ.) IBISML2014-35 |
We apply partial parallel interference cancellation scheme in CDMA demodulation to sparse signal recovery in compressed ... [more] |
IBISML2014-35 pp.1-7 |
IBISML |
2014-11-17 17:00 |
Aichi |
Nagoya Univ. |
[Poster Presentation]
Efficient Method for Computing Belief Propagation, with Application to CDMA Multiuser Detection Arise Kuriya, Toshiyuki Tanaka (Kyoto Univ) IBISML2014-54 |
Pearl's belief propagation (BP) is an algorithm to solve
inference problems on probability models defined in terms of
... [more] |
IBISML2014-54 pp.149-153 |