Committee |
Date Time |
Place |
Paper Title / Authors |
Abstract |
Paper # |
IBISML |
2022-03-08 10:25 |
Online |
Online |
Robust computation of optimal transport by β-potential regularization Shintaro Nakamura (Univ. Tokyo), Han Bao (Univ.Tokyo/RIKEN), Masashi Sugiyama (RIKEN/Univ. Tokyo) IBISML2021-31 |
Optimal transport (OT) has become a widely used tool to measure the discrepancy between probability distributions
in th... [more] |
IBISML2021-31 pp.8-14 |
IBISML |
2022-03-09 13:30 |
Online |
Online |
Is the Performance of My Deep Network Too Good to Be True?
-- A Direct Approach to Estimating the Bayes Error in Binary Classification -- Takashi Ishida (UTokyo), Ikko Yamane (Université Paris Dauphine-PSL/RIKEN), Nontawat Charoenphakdee (UTokyo), Gang Niu (RIKEN), Masashi Sugiyama (RIKEN/UTokyo) IBISML2021-44 |
[more] |
IBISML2021-44 pp.38-45 |
IBISML |
2021-03-02 10:00 |
Online |
Online |
Learning from Noisy Complementary Labels with Robust Loss Functions Hiroki Ishiguro (UTokyo), Takashi Ishida (UTokyo/RIKEN), Masashi Sugiyama (RIKEN/UTokyo) IBISML2020-34 |
It has been demonstrated that large-scale labeled datasets facilitate the success of machine learning. However, collecti... [more] |
IBISML2020-34 pp.1-8 |
IBISML |
2020-10-20 13:20 |
Online |
Online |
[Invited Talk]
* IBISML2020-12 |
(To be available after the conference date) [more] |
IBISML2020-12 p.28 |
IBISML |
2017-11-09 13:00 |
Tokyo |
Univ. of Tokyo |
Semi-Supervised AUC Optimization based on Positive-Unlabeled Learning Tomoya Sakai, Gang Niu (UTokyo/RIKEN), Masashi Sugiyama (RIKEN/UTokyo) IBISML2017-40 |
Maximizing the area under the receiver operating characteristic curve (AUC) is a standard approach to imbalanced classif... [more] |
IBISML2017-40 pp.39-46 |
IBISML |
2017-11-10 13:00 |
Tokyo |
Univ. of Tokyo |
[Poster Presentation]
Multi-Task Learning with Positive and Unlabeled Data and Its Application to Mental State Prediction Hirotaka Kaji, Hayato Yamaguchi (Toyota Motor), Masashi Sugiyama (RIKEN/UTokyo) IBISML2017-66 |
In real-world machine learning applications, we are often faced with a situation where only a small number of training s... [more] |
IBISML2017-66 pp.235-242 |
NC, IPSJ-BIO, IBISML, IPSJ-MPS [detail] |
2017-06-24 10:20 |
Okinawa |
Okinawa Institute of Science and Technology |
Risk Minimization Framework for Multiple Instance Learning from Positive and Unlabeled Bags Han Bao (Univ. of Tokyo), Tomoya Sakai, Issei Sato (Univ. of Tokyo/RIKEN), Masashi Sugiyama (RIKEN/Univ. of Tokyo) IBISML2017-3 |
Multiple instance learning (MIL) is a variation of traditional supervised learning problems where data (referred to as b... [more] |
IBISML2017-3 pp.55-62 |
NC, IPSJ-BIO, IBISML, IPSJ-MPS [detail] |
2017-06-24 10:45 |
Okinawa |
Okinawa Institute of Science and Technology |
Positive-Unlabeled Learning with Non-Negative Risk Estimator Ryuichi Kiryo (Univ. of Tokyo/RIKEN), Gang Niu (Univ. of Tokyo), Masashi Sugiyama (RIKEN/Univ. of Tokyo) IBISML2017-4 |
From only emph{positive}~(P) and emph{unlabeled}~(U) data, a binary classifier can be trained with PU learning, in which... [more] |
IBISML2017-4 pp.63-70 |
NC, IPSJ-BIO, IBISML, IPSJ-MPS [detail] |
2017-06-24 11:10 |
Okinawa |
Okinawa Institute of Science and Technology |
Learning from Complementary Labels Takashi Ishida (SMAM/Univ. of Tokyo), Gang Niu (Univ. of Tokyo), Masashi Sugiyama (RIKEN/Univ. of Tokyo) IBISML2017-5 |
[more] |
IBISML2017-5 pp.71-78 |
NC, IPSJ-BIO, IBISML, IPSJ-MPS [detail] |
2017-06-25 09:30 |
Okinawa |
Okinawa Institute of Science and Technology |
Expectation Propagation for t-Exponential Family Futoshi Futami, Issei Sato (Univ. of Tokyo/RIKEN), Masashi Sugiyama (RIKEN/Univ. of Tokyo) IBISML2017-6 |
Exponential family distributions are highly useful in machine learning since their calculation can be performed efficien... [more] |
IBISML2017-6 pp.179-184 |
NC, IPSJ-BIO, IBISML, IPSJ-MPS [detail] |
2017-06-25 09:55 |
Okinawa |
Okinawa Institute of Science and Technology |
Stochastic Divergence Minimization for Biterm Topic Model Zhenghang Cui (Univ. of Tokyo), Issei Sato (Univ. of Tokyo/RIKEN), Masashi Sugiyama (RIKEN/Univ. of Tokyo) IBISML2017-7 |
Inferring latent topics of collected short texts is useful for understanding its hidden structure and predicting new con... [more] |
IBISML2017-7 pp.185-192 |
IBISML |
2016-11-17 14:00 |
Kyoto |
Kyoto Univ. |
Semi-Supervised Classification Based on Classification from Positive and Unlabeled Data Tomoya Sakai, Marthinus Christoffel du Plessis, Gang Niu (UTokyo), Masashi Sugiyama (RIKEN/UTokyo) IBISML2016-80 |
Most of the semi-supervised learning methods developed so far use unlabeled data for regularization purposes under parti... [more] |
IBISML2016-80 pp.243-250 |
IBISML |
2016-03-17 13:50 |
Tokyo |
Institute of Statistical Mathematics |
Least-Squares Log-Density Gradient Clustering for Riemannian Manifolds Mina Ashizawa (UTokyo), Hiroaki Sasaki (NAIST), Tomoya Sakai, Masashi Sugiyama (UTokyo) IBISML2015-96 |
[more] |
IBISML2015-96 pp.17-24 |
IBISML |
2015-11-26 15:00 |
Ibaraki |
Epochal Tsukuba |
[Poster Presentation]
Geometry-Aware Stationary Subspace Analysis for Multivariate Signals Inbal Horev (Tokyo Tech), Florian Yger (Universite Paris-Dauphine), Masashi Sugiyama (UTokyo) IBISML2015-67 |
[more] |
IBISML2015-67 pp.107-114 |
IBISML |
2015-11-27 14:00 |
Ibaraki |
Epochal Tsukuba |
[Poster Presentation]
Sufficient Dimension Reduction via Direct Estimation of the Gradients of Logarithmic Conditional Densities Hiroaki Sasaki, Voot Tangkaratt, Masashi Sugiyama (UTokyo) IBISML2015-81 |
[more] |
IBISML2015-81 pp.209-216 |
IBISML |
2015-11-27 14:00 |
Ibaraki |
Epochal Tsukuba |
[Poster Presentation]
Non-Gaussian Component Analysis with Log-Density-Gradient Estimation Hiroaki Sasaki, Gang Niu, Masashi Sugiyama (UTokyo) IBISML2015-82 |
[more] |
IBISML2015-82 pp.217-224 |
IBISML |
2014-11-17 17:00 |
Aichi |
Nagoya Univ. |
[Poster Presentation]
Theoretical Analysis of Empirical MAP and Empirical Partially Bayes Shinichi Nakajima (TU Berlin), Masashi Sugiyama (Univ. of Tokyo) IBISML2014-38 |
Variational Bayesian (VB) learning is known to be a
promising
approximation to Bayesian learning
with computational... [more] |
IBISML2014-38 pp.25-32 |
IBISML |
2014-11-17 17:00 |
Aichi |
Nagoya Univ. |
[Poster Presentation]
Multitask learning meets tensor factorization: task imputation via convex optimization Kishan Wimalawarne (Tokyo Inst. of Tech.), Masashi Sugiyama (Univ. of Tokyo), Ryota Tomioka (TTIC) IBISML2014-49 |
We study a multitask learning problem in which each task is parametrized by a weight vector and indexed by a pair of ind... [more] |
IBISML2014-49 pp.111-118 |
IBISML |
2014-11-17 17:00 |
Aichi |
Nagoya Univ. |
[Poster Presentation]
Direct Density Ratio Estimation with Convolutional Neural Networks with Application in Outlier Detection Hyunha Nam (Tokyo Inst. of Tech.), Masashi Sugiyama (Univ. of Tokyo) IBISML2014-51 |
(Advance abstract in Japanese is available) [more] |
IBISML2014-51 pp.127-132 |
IBISML |
2014-11-17 17:00 |
Aichi |
Nagoya Univ. |
[Poster Presentation]
Target Shift Adaptation for Conditional Density Estimation and Regression Tuan Duong Nguyen (Tokyo Inst. of Tech.), Marthinus Christoffel du Plessis, Masashi Sugiyama (Univ. of Tokyo) IBISML2014-55 |
We consider the problem of domain adaptation under target shift scenario, where the target marginal distributions p(y) d... [more] |
IBISML2014-55 pp.155-160 |