News

Bionsight, inc. is a fast growing bio-medicine company driven by strong AI technologies.

The interview with Pharm Edaily has been released at 2022.07.06. It can be found here (in Korean): https://pharm.edaily.co.kr/news/read?newsId=01262806632391896

I will present our paper at the Bernoulli-IMS 10th World Congress in Probability and Statistics, 19-23 July 2021, Seoul National University, Seoul, South Korea.

The congress will be held virtually based on Zoom format.

The title of the presentation is "Analysis of chemical-gene bipartite network via a user-based collaborative filtering method incorporating chemical structure information".

The paper is a collaborative work with Bionsight, Inc.

I presented our paper at the 2019 spring conference of the Korean Statistical Society, 24-25 May, 2019,

which was held at Kangwon National University, Chuncheon, Gangwon, South Korea.

The title of the presentation was "Analysis of Brain Functional Connectivity Based on Copula Directional Dependence".

The paper is a collaborative work with Prof. Jong-Min Kim at the University of Minnesota, USA.

I presented our work on Block Tensor Train Decomposition for Missing Data Imputation

at the Asia-Pacific Signal and Information Processing Association Annual Summit and Conference 2018 (APSIPA ASC 2018)

November 12-15, 2018, Hawaii Convention Center, Honolulu, Hawaii, USA (www.apsipa2018.org)

I have presented my work on

"Shrinkage estimation and model selection for vector autoregressive models with an application to fMRI data analysis"

at the Pusan National University, Busan, South Korea.

for the 2018 Spring Conference of the Korean Statistical Society.

This is a joint work with Prof. S.-H. Kim (KAIST)

We are also building an R package related to this work!

I have presented my work on

"Block tensor train decomposition for missing data estimation"

at Polytechnic University of Catalonia, Barcelona, Spain.

joint work with Prof. J.-M. Kim (U. Minnesota-Morris).

We are doing a very productive and great collaboration on multidimensional complex data analysis and statistical modeling.

I have presented our work on

"Robust multilinear rank estimation for Tucker tensor regression"

at City University of Hong Kong.

joint work with Prof. T. Yokota (Nagoya Institute of Technology) and Prof. J.-M. Kim (U. Minnesota-Morris).

We are currently doing a great collaboration for tensor-based data mining and many real applications.

Recently I joined the Department of Information Statistics, Kangwon National University, Chuncheon, Republic of Korea.

Chuncheon is a central city of Gangwon province, and it is located near Seoul.

It takes one-hour by ITX train from the center of Seoul.

My previous RIKEN email address is still valid at least until March 2018.

Our paper "Nonnegative Tensor Train Decompositions for Multi-Domain Feature Extraction and Clustering"

received the "Excellent Paper Award"

at the ICONIP 2016, 16-21 October 2016, Kyoto, Japan.

This is a joint work with Drs. Anh-Huy Phan, Fengyu Cong, Andrzej Cichocki.

The co-organizers of the ICONIP 2016 are: APNNS (Asia Pacific Neural Network Society) and JNNS (Japanese Neural Network Society).

I presented our work at the 23rd International Conference on Neural Information Processing (ICONIP2016).

Our title of the presentation is "Nonnegative Tensor Train Decompositions for Multi-Domain Feature Extraction and Clustering".

This is a joint work with Drs. A.-H. Phan, F. Cong, A. Cichocki.

The Korean Statistical Society Spring Conference 2016 was held at Kyungpook National University (KNU), Daegu, South Korea.

http://www.kss.or.kr/bbs/board.php?bo_table=scheduleview&idx=116&mode=last

I gave a presentation on a multilinear rank selection method for multi-way array data analysis.

Workshop on Tensor Decompositions and Applications (TDA 2016) was held in the Leuven Institute of Ireland (the Irish College), Leuven, Belgium.

http://www.esat.kuleuven.be/stadius/TDA2016/index.php

We are grateful to the organizers for organizing the great workshop and giving us opportunity to present our research.

The title of the presentation: Tensor Train Decompositions for Higher Order Regression with LASSO Penalties.

joint work with Professor Andrzej Cichocki, RIKEN Brain Science Institute, Japan.

Workshop "Low-rank Optimization and Applications" was held by Hausdorff Center for Mathematics (HCM).

http://www.hcm.uni-bonn.de/low-rank-optimization-2015

We are very grateful to the organizers and HCM for invitation, organization, and financial support.

The title of the presentation: Low-Rank Tensor Train/Tensor Networks for Large-Scale Optimization Problems,

joint work with Professor Andrzej Cichocki, RIKEN Brain Science Institute, Japan.

I presented our work at the 11th International Symposium on Neural Networks (ISNN 2014).

The title of the presentation is "Very Larage-Scale Singular Value Decomposition Using Tensor Train Networks".

The study was carried out in the Cichocki Lab. for Advanced Brain Signal Processing, RIKEN Brain Science Institute, Japan.