| Technical Committee on Information-Based Induction Sciences and Machine Learning (IBISML) |
[schedule] [select]
|
|
| Chair |
|
Toshihiro Kamishima (independent researcher) |
| Vice Chair |
|
Atsuyoshi Nakamura (Hokkaido Univ.), Koji Tsuda (Univ. of Tokyo) |
| Secretary |
|
Yoshinobu Kawahara (Osaka Univ.), Taiji Suzuki (Univ.of Tokyo) |
| Assistant |
|
Mahito Sugiyama (NII), SAKUMA JUN (Tokyo Tech) |
|
| Conference Date |
Fri, Dec 20, 2024 09:00 - 17:40
Sat, Dec 21, 2024 09:00 - 17:20 |
| Topics |
generative AI, machine learning |
| Conference Place |
Lecture Room 1, Graduate School of Environmental Science |
| Address |
N10W5, Kita-ku, Sapporo, Hokkaido 060-0810 |
| Transportation Guide |
10min walk from JR Sapporo st., or 5min walk from a subway st. “Kita 12-jou”
https://www.hokudai.ac.jp/bureau/property/ees/access/ |
Contact
Person |
Prof. Atsuyoshi Nakamura |
| Registration Fee |
This workshop will be held as the IEICE workshop in fully electronic publishing. Registration fee will be necessary except the speakers and participants other than the participants to workshop(s) in non-electronic publishing. See the registration fee page. We request the registration fee or presentation fee to participants who will attend the workshop(s) on IBISML. |
| Due for Registration |
Please proceed the payment of registration fee by 3 days before the workshop date. The meeting URL will be sent from one of the secretaries of the committee via e-mail, just before the workshop date. |
Fri, Dec 20 AM
09:00 - 10:20 |
| (1) |
09:00-09:20 |
Multi-Objective Bayesian Optimization via Maximizing a Variational Lower Bound on Mutual Information |
Masanori Ishikura, Masayuki Karasuyama (NITech) |
| (2) |
09:20-09:40 |
Combinatorial Pure Exploration for Correlation Clustering with Noisy Oracle |
Yuko Kuroki, Atsushi Miyauchi (CENTAI), Francesco Bonchi (CENTAI/Eurecat), Wei Chen (MSR) |
| (3) |
09:40-10:00 |
Q-Learning with Prior Knowledge |
Takahisa Imagawa, Shuichi Enokida (KIT) |
| (4) |
10:00-10:20 |
Train marshaling plan considering weight of freight cars
-- A design method using reinforcement learning -- |
Yoichi Hirashima (OIT) |
| |
10:20-10:30 |
Break ( 10 min. ) |
Fri, Dec 20 AM
10:30 - 11:30 |
(5)
[updated] |
10:30-10:50 |
Training Deep Neural Networks for Fast Data Assimilation |
Yuta Ono (UTokyo), Yuta Tarumi, Keisuke Fukuda, Shin-ichi Maeda (PFN) |
(6)
[updated] |
10:50-11:10 |
Active learning considering test distribution with Gaussian process regression model |
Yoshito Okura (Nagoya Univ.), Shion Takeno (Nagoya Univ./RIKEN), Yu Inatsu (NITech), Tatsuya Aoyama, Tomonari Tanaka, Satoshi Akahane (Nagoya Univ.), Hiroyuki Hanada, Noriaki Hashimoto (RIKEN), Taro Murayama, Lee Hanju, Shinya Kojima (DENSO), Ichiro Takeuchi (Nagoya Univ./RIKEN) |
(7)
[updated] |
11:10-11:30 |
Bayesian modeling of spectral decomposition process by scientist |
Hayato Okubo, Yoshifumi Amamoto, Hiroyuki Kumazoe, Toshimitsu Aritake, Youichi Mototake (Hitotsubashi Univ.) |
| |
11:30-13:00 |
Lunch Break ( 90 min. ) |
Fri, Dec 20 PM
13:00 - 14:20 |
(8)
[updated] |
13:00-13:20 |
Probability density estimation |
Masahito Toba, Seiichi Uchida (Kyushu Univ.), Hideaki Hayashi (Osaka Univ.) |
(9)
[updated] |
13:20-13:40 |
Generative Inference for object poses by Regressive-feature Reconstruction |
Ryo Isobe, Rei Kawakami (Science Tokyo), Satoshi Ikehata (Science Tokyo/NII), Ikuro Sato (Science Tokyo/Denso IT Lab) |
(10)
[updated] |
13:40-14:00 |
Mechanism of Shape Development of Symbolic Spectra of Amorphous Glasses by Topological Data Analysis |
Hidetoshi Morita, Youichi Mototake, Yoshihumi Amamoto, Hayato Okubo, Naoki Tosa (Hitotsubashi Univ.) |
(11)
[updated] |
14:00-14:20 |
Conformal Prediction and Bayesian Optimization
[updated] |
Shun Nanjo (SOKENDAI), Ryo Yoshida (ISM) |
| |
14:20-14:30 |
Break ( 10 min. ) |
Fri, Dec 20 PM
14:30 - 16:30 |
(12)
[updated] |
14:30-15:10 |
|
(13)
[updated] |
15:10-15:50 |
|
(14)
[updated] |
15:50-16:30 |
|
| |
16:30-16:40 |
Break ( 10 min. ) |
Fri, Dec 20 PM
16:40 - 17:40 |
(15)
[updated] |
16:40-17:00 |
Information geometry of nonmonotonic quantum natural gradient |
Hideyuki Miyahara (Hokkaido Univ.) |
(16)
[updated] |
17:00-17:20 |
Bayesian Optimization for Simultaneous Selection of Machine Learning Algorithms and Hyperparameters on Shared Latent Space |
Kazuki Ishikawa, Ryota Ozaki, Yohei Kanzaki (NITech), Ichiro Takeuchi (Nagoya Univ/RIKEN), Masayuki Karasuyama (NITech)
[updated] |
(17)
[updated] |
17:20-17:40 |
Revising Rulelists via Integer Linear Programming |
Yoichi Sasaki, Yuzuru Okajima (NEC), Hiroki Arimura (Hokkaido University) |
Sat, Dec 21 AM
09:00 - 11:00 |
(18)
[updated] |
09:00-09:40 |
|
(19)
[updated] |
09:40-10:20 |
|
(20)
[updated] |
10:20-11:00 |
|
| |
11:00-11:10 |
Break ( 10 min. ) |
Sat, Dec 21 AM
11:10 - 11:50 [updated] |
(21)
[updated] |
11:10-11:30 |
Selective Inference for Auto Feature Engineering |
Tatsuya Matsukawa, Tomohiro Shiraishi (Nagoya Univ.), Shuichi Nishino (Nagoya Univ./RIKEN), Teruyuki Katsuoka (Nagoya Univ.), Ichiro Takeuchi (Nagoya Univ./RIKEN) |
(22)
[updated] |
11:30-11:50 |
Assessing predictive performance and model interpretability through SHAP-based feature selection |
Ryota Hashiura, Akihiro Omori, Naoto Nakano (Meiji Univ.) |
| |
11:50-13:40
[updated] |
Lunch Break ( 110 min. )
[updated] |
Sat, Dec 21 PM
13:40 - 15:00 |
(23)
[updated] |
13:40-14:00 |
Global Optimization for Empirical Risk Function by Graduated Optimization with Stochastic Noise in SGD
[updated] |
Naoki Sato, Hideaki Iiduka (Meiji Univ.) |
(24)
[updated] |
14:00-14:20 |
Adversarial Robustness by Mixed Features of L1 regularized Neural Networks
[updated] |
Hinata Inaoka, Yoshinari Takeishi, Jun'ichi Takeuchi (Kyushu Univ.)
[updated] |
(25)
[updated] |
14:20-14:40 |
Singularities in a Learning Machine: Estimating the global learning coefficient in the transformers |
Yusuke Hayashi (ALIGN), Ryosuke Sakamoto (HU), Kotaro Sakamoto (UT) |
(26)
[updated] |
14:40-15:00 |
First numerical observation of the Berezinskii-Kosterlitz-Thouless transition in language models
[updated] |
Yuma Toji (Hokkaido Univ.), Jun Takahashi (UTokyo), Vwani Roychowdhury (UCLA), Hideyuki Miyahara (Hokkaido Univ.) |
| |
15:00-15:10 |
Break ( 10 min. ) |
Sat, Dec 21 PM
15:10 - 16:30 |
(27)
[updated] |
15:10-15:30 |
Unrolling Nonnegative Matrix Factorization via Learning a Single Optimization Step
[updated] |
Toshimitsu Aritake (Hitotsubashi Univ.) |
(28)
[updated] |
15:30-15:50 |
Distributionally Robust Training Instances Selection with Guaranteed Model Performance
[updated] |
Tomonari Tanaka (Nagoya Univ.), Hiroyuki Hanada (RIKEN), Hanting Yang (Nagoya University), Tatsuya Aoyama (Nagoya Univ.), Yu Inatsu (NITech), Satoshi Akahane, Yoshito Okura (Nagoya Univ.), Noriaki Hashimoto (RIKEN), Taro Murayama, Lee Hanju, Shinya Kojima (DENSO), Ichiro Takeuchi (Nagoya Univ./RIKEN) |
(29)
[updated] |
15:50-16:10 |
Evaluation of exponentially weighted average algorithm using quantile regression for non-stationary time series
[updated] |
Akihiro Omori, Ryota Hashiura, Naoto Nakano (Meiji Univ.) |
(30)
[updated] |
16:10-16:30 |
Concept Drift Detection Using Unupdated Autoencoder |
Taisei Takano, Hisashi Koga (UEC) |
| |
16:30-16:40 |
Break ( 10 min. ) |
Sat, Dec 21 PM
16:40 - 17:20 |
(31)
[updated] |
16:40-17:00 |
Neural Network-Based Melody Generation in Low-Resource Environments with Rule-Based Enhancements: Methodology and Evaluation. |
Masaki Matsunaga, Lovic Gauthier (Ariake kosen) |
(32)
[updated] |
17:00-17:20 |
Classifying information about childcare using machine learning |
MayinuerZipaer (SERIO) |
| Announcement for Speakers |
| General Talk | Each speech will have 15 minutes for presentation and 5 minutes for discussion. |
| Contact Address and Latest Schedule Information |
| IBISML |
Technical Committee on Information-Based Induction Sciences and Machine Learning (IBISML) [Latest Schedule]
|
| Contact Address |
Atsuyoshi Nakamura (Hokkaido University) a  ist    i |
Last modified: 2024-11-28 14:00:19
|