グローバルイノベーション研究院

お問い合わせ

アイコン

メニュー

グローバルイノベーション研究院

English

イベント情報

ネット配信【GIR公開セミナー】 Dr. Anh-Huy Phan / スコルコボ科学技術大学(ロシア)”Recent Advances in Signal Processing and Machine Learning”

日時 2020.2.27(16:30~18:00)
会場 東京農工大学 小金井キャンパス BASE会議室
講演者・講演タイトル 以下のリンク、または電話番号からセミナーに参加できます。
—-
https://zoom.us/j/655497359

ミーティングID: 655 497 359

所在地でダイアル
+1 929 205 6099 アメリカ合衆国 (ニューヨーク)
+1 669 900 6833 アメリカ合衆国 (サンノゼ)
+81 524 564 439 日本
+81 3 4578 1488 日本
ミーティングID: 655 497 359
------------------------------------------------------------------------------------------------------------------------------
『Recent Advances in Signal Processing and Machine Learning』

16:30~17:00
◆Dr. Anh-Huy Phan (Assistant Professor, Center for Computational and Data-Intensive Science and Engineering, Skolkovo Institute of Science and Technology, Russia)
 
 "Novel Algorithms for Tensor-train approximation"

〈要旨〉
Decompositions of tensors into factor matrices which interact through a core tensor, have found numerous applications in signal processing and machine learning.
A more general tensor model that represents data as an ordered network of subtensors of order-2 or order-3 has, so far, not been widely considered in these fields,
although this so-called tensor network (TN) decomposition has been long studied in quantum physics and scientific computing.
In this article, we present novel algorithms and applications of TN decompositions, with a particular focus on the tensor train (TT) decomposition and its variants.
The novel algorithms developed for the TT decomposition update, in an alternating way, one or several core tensors at each iteration and exhibit enhanced mathematical
tractability and scalability for large-scale data tensors. For rigor, the cases of the given ranks, given approximation error, and the given error bound are all considered.
The proposed algorithms provide well-balanced TT-decompositions and are tested in the classic paradigms of blind source separation from a single mixture,
denoising, and feature extraction, achieving superior performance over the widely used truncated algorithms for TT decomposition.


17:00~17:30
◆Dr. Simone Fiori (Associate professor, Università Politecnica delle Marche, Italy)
 
 "Virtual attractive-repulsive potentials extended to manifolds"

〈要旨〉
In this talk I will summarize a non-linear control theory from
a scientific work published in 2005 by Caltech researchers ("Virtual attractive-repulsive potentials for cooperative control of second order dynamic vehicles") and how it was used to control a small robot.
Further, I will show how such control theory may be extended to smooth manifolds and, at the end, I will mention how my students and I are trying to employ this theory to achieve self-guidance of an autonomous quadcopter drone.


17:30~18:00
◆Md. Khademul Islam Molla (Professor, Computer Science and Engineering, University of Rajshahi, Bangladesh)

 "Frequency recognition of short-time SSVEP signal in BCI paradigm using multistage parameter fusion"

〈要旨〉
Brain-computer interface (BCI) refers to the recognition of brain activity leading to generate corresponding commands to interact with external devices. Due to its safety and high time resolution, electroencephalogram (EEG) based BCIs have become popular. Steady-state visual evoked potential (SSVEP) is an EEG particularly attractive due to high signal to noise ratio and robustness. Reducing calibration time in SSVEP is a demanding issue in the related research community.
A multistage parameter fusion-based approached is studied to recognize the frequency of short-time SSVEP for brain-computer interface (BCI) implementation. Two reference signals are generated by averaging the training trial considering a leave-one-out (LOO) method. The signal of each channel is passed through a filterbank designed to cover all the stimulus frequencies as well as their harmonics.
A noticeable result is obtained using the method although further improvement is required. We also study on the improvement of frequency recognition with short-time SSVEP using data augmentation.
言語 英語
対象 どなたでも、ご聴講いただけます。
共催 JST CREST 人工知能 「てんかんAI」田中プロジェクト
グローバルイノベーション研究院 ライフサイエンス分野 田中チーム
卓越大学院プログラム 
開催概要
お問い合わせ窓口 グローバルイノベーション研究院・工学研究院 田中聡久
e-mail: tanakat (ここに@ を入れてください) cc.tuat.ac.jp

COPYRIGHT 2020 ©TOKYO UNIVERSITY OF AGRICULTURE AND TECHNOLOGY., ALL RIGHTS RESERVED.

このページの上部へ

メニューを閉じる