Invited Talk (10:00 AM JST, September 26, 2023)
Title: Making the Invisible Visible: Toward High-Quality Deep THz Computational Imaging
Abstract: Terahertz (THz) computational imaging has recently attracted significant attention thanks to its non-invasive, non-destructive, non-ionizing, material-classification, and ultra-fast nature for 3D object exploration and inspection. However, its strong water absorption nature and low noise tolerance lead to undesired blurs and distortions of reconstructed THz images. The performances of existing methods are highly constrained by the diffraction-limited THz signals. In this talk, we will introduce the characteristics of THz imaging and its applications. We will also show how to break the limitations of THz imaging with the aid of complementary information between the THz amplitude and phase images sampled at prominent frequencies (i.e., the water absorption profile of THz signal) for THz image restoration. To this end, we propose a novel physics-guided deep neural network design, namely Subspace-Attention-guided Restoration Network (SARNet), that fuses such multi-spectral features of THz images for effective restoration. Furthermore, we experimentally construct an ultra-fast THz time-domain spectroscopy system covering a broad frequency range from 0.1 THz to 4 THz for building up temporal/spectral/spatial/phase/material THz database of hidden 3D objects.
Bio: Prof. Chia-Wen Lin is currently a Professor with the Department of Electrical Engineering, National Tsing Hua University (NTHU), Taiwan. He also serves as Deputy Director of the AI Research Center of NTHU. He is currently a Visiting Professor at the Graduate School of Informatics, Informatics, Kyoto University from July 2023 to December 2023. He served as Visiting Professor at Nagoya University and National Institute of Informatics, Japan, in 2019 and 2015, respectively. His research interests include image/video processing, computer vision, and video networking.
Dr. Lin is an IEEE Fellow, and has been serving on IEEE Circuits and Systems Society (CASS) Fellow Evaluating Committee since 2021. He serves as IEEE CASS BoG member-at-Large during 2022-2024. He was Steering Committee Chair of IEEE ICME (2020-2021), IEEE CASS Distinguished Lecturer (2018-2019), and President of the Chinese Image Processing and Pattern Recognition (IPPR) Association, Taiwan (2019-2020). He has served as Associate Editor of IEEE Transactions on Image Processing, IEEE Transactions on Multimedia, IEEE Transactions on Circuits and Systems for Video Technology, and IEEE Multimedia. He served as TPC Chair of IEEE ICME in 2010 and IEEE ICIP in 2019, and the Conference Chair of IEEE VCIP in 2018.
Registration link: https://zoom.us/meeting/register/tJAld-ypqj8rGtf5dFOnsJ8D_cx-10yDBZAs
Invited Talk (1:00–1:45 PM JST, March 1, 2023)
Title: What Do Self-Supervised Speech Representation Models Know? A Layer-Wise Analysis
This Invited Talk was provided as a part of SPEASIP.
Special Invited Talk (1:00–2:00 PM JST, December 14, 2022)
第37回信号処理シンポジウムにおいて特別招待講演を企画しました.詳細については信号処理シンポジウムのページをご覧ください.なお,本講演は信号処理シンポジウムの参加者のみご参加頂けます.
Seminar (4:50–5:50 PM JST, June 16, 2022)
Title: Graph Constructions for Machine Learning Applications: New Insights and Algorithms
Abstract: Graphs have long been used in a wide variety of problems, such as in analysis of social networks, machine learning, network protocol optimization or image processing. In the last few years, a growing body of work has been developed to extend and complement well known concepts in spectral graph theory, leading to the emergence of Graph Signal Processing (GSP) as a broad research field. In this talk we focus on summarizing recent results that lead to a GSP perspective of machine learning problems. The key observation is that representations of sample data points (e.g., images in a training set) can be used to construct graphs, with nodes representing samples, label information resulting in graph signals, and edge weights capturing the relative positions of samples in feature space. We will first review how this perspective has been used in well known techniques for label propagation and semi-supervised learning. Then, we will introduce the non-negative kernel regression (NNK) graph construction, describe its properties, and introduce example applications in machine learning areas such as i) model explainability, ii) local interpolative classification and iii) self-supervised learning.
Bio: Antonio Ortega is a professor of Electrical and Computer Engineering at the University of Southern California (USC). He received his undergraduate and doctoral degrees from the Universidad Politecnica de Madrid, Madrid, Spain and Columbia University, New York, NY, respectively. He is a Fellow of the IEEE and EURASIP, and a member of ACM and APSIPA. He was the Editor-in-Chief of the IEEE Transactions of Signal and Information Processing over Networks and recently served as a member of the Board of Governors of the IEEE Signal Processing Society. He has received several paper awards, including the 2016 Signal Processing Magazine award. His recent research work is focusing on graph signal processing, machine learning, multimedia compression and wireless sensor networks. He is the author of the book, "Introduction to Graph Signal Processing", published by Cambridge University Press in 2022.
Workshop (9:20–10:10 AM JST, March 2, 2022)
We will have a workshop at 9:20-10:10 AM JST, March 2, 2022. We will invite Prof. Masahiro Yukawa (Keio University, JAPAN) as a lecturer. This workshop will be held in a hybrid manner (onsite (Naha, Okinawa) and online (Zoom)). For onsite participation, please refer to IEICE web page. For online participation, please register your name and e-mail address here. Anyone can join for free online. If you have any questions, contact us (higashi-h [at] i.kyoto-u.ac.jp).
湯川正裕先生 (慶應義塾大学) の招待講演をオンサイト(沖縄県立博物館・美術館)とオンライン(Zoom)のハイブリッド形式で,2022年3月2日午前9時20分〜10時10分で開催します.オンサイトでの参加はこちらをご参照ください.オンラインでの参加はこちらから名前とメールアドレスをご登録ください.無料でどなたでも聴講できます.ご質問があれば higashi-h [at] i.kyoto-u.ac.jp までご連絡ください.
Onsite: Registration for on-site participation will be mandatory. Details will be announced on IEICE Web Page. 現地参加には登録が必要です.詳しくはIEICEのページをご覧ください.
Online: Register your name and e-mail address here
Title: Be robust against outlier, and be stable under high-power Gaussian noise simultaneously
Abstract: Outlier robustness is ubiquitous in signal processing and machine learning. Although significant amount of efforts have been devoted to enhance robustness, the existing methods suffer from some limitations. In this talk, I will introduce our recent studies on robust regression which have two key components. The first is the adoption of the minimax concave (MC) loss, leveraging the idea of the MC penalty for enhancing robustness by sparsifying the outlier estimate. Since the gradient of the MC loss vanishes at a certain thresholding level, it is highly robust even in the presence of catastrophic outliers, as opposed to the well-known Huber loss which is relatively sensitive to such large outliers compared to nonconvex losses. At the same time, it is mathematically tractable due to its weak convexity which permits convexity of the entire cost function, unlike the classical (and nonconvex) Tukey loss. The second component is the introduction of auxiliary vector to model the Gaussian noise. This allows to reflect the noise Gaussianity and the outlier sparsity in a reasonable manner, leading to remarkable robustness and stability in the presence of huge outlier and severe Gaussian noise. Under those key components, I will present a mathematically rigorous framework as well as some simulation results showing the remarkable performance of the proposed approach.
Bio: Masahiro Yukawa received the B.E., M.E., and Ph.D. degrees from the Tokyo Institute of Technology, Tokyo, Japan, in 2002, 2004, and 2006, respectively. He was a Visiting Researcher/Professor with the University of York, U.K. (October 2006 -- March 2007), with the Technical University of Munich, Germany (July 2008 -- November 2008), and with the Technical University of Berlin, Germany (April 2016 -- February 2017). He was a Special Postdoctoral Researcher with RIKEN, Japan (2007 -- 2010), and an Associate Professor with Niigata University, Japan (2010 -- 2013). He is currently an Associate Professor with the Department of Electronics and Electrical Engineering, Keio University, Yokohama, Japan. He was also a Visiting Scientist with AIP Center, RIKEN, Wakou, Japan (July 2017 -- March 2020). His research interests include mathematical adaptive signal processing, convex/sparse optimization, and machine learning. He was an Associate Editor for the IEEE TRANSACTIONS ON SIGNAL PROCESSING (2015 -- 2019), Multidimensional Systems and Signal Processing (2012 -- 2016), and the IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences (2009–2013). He was the recipient of the Research Fellowship of the Japan Society for the Promotion of Science from April 2005 to March 2007. He was also the recipient of the Excellent Paper Award and the Young Researcher Award from the IEICE in 2006 and 2010, respectively, the Yasujiro Niwa Outstanding Paper Award in 2007, the Ericsson Young Scientist Award in 2009, the TELECOM System Technology Award in 2014, the Young Scientists Prize, the Commendation for Science and Technology by the Minister of Education, Culture, Sports, Science and Technology in 2014, the KDDI Foundation Research Award in 2015, the FFIT Academic Award in 2016, and the JSPS Prize in 2021. He is a Senior Member of IEEE and a Member of the Institute of Electronics, Information and Communication Engineers.
電子情報通信学会 信号処理研究会 Technical Committee of Signal Processing (SIP), Institute of Electronics, Information and Communication Engineers (IEICE)
電子情報通信学会および日本音響学会 音声研究会 Technical Committee of Speech (SP), Institute of Electronics, Information and Communication Engineers (IEICE)
電子情報通信学会 応用音響研究会 および日本音響学会 電気音響研究会 Technical Committee of Engineering acoustics (EA), Institute of Electronics, Information and Communication Engineers (IEICE)
情報処理学会 音声言語情報処理研究会 Special Interest Group on Spoken Language Processing (SLP), Information Processing Society of Japan (IPSJ)
IEEE Signal Processing Society Kansai Chapter
Seminar (11:00–12:00 AM JST, February 24, 2022)
Institute of Global Innovation Research (GIR), Tokyo University of Technology (TUAT), and APSIPA Japan Chapter will organize a seminar at 11:00-12:10 AM JST, February 24, 2022. We will invite Prof. Gene Cheung (York University, Canada) as a lecturer. This workshop will be held on Zoom. Anyone can join for free. Please register your name and e-mail address here. For the detail, visit the webpage.
東京農工大学グローバルイノベーション研究院とAPASIPA Japan Capterは,Gene Cheung 先生 (York University, Canada) の招待講演を2022年2月24日午前11時00分〜12時00分で開催します.無料でどなたでも聴講できます.こちらから名前とメールアドレスをご登録ください.詳しくはこちらのページをご参照ください.
Register your name and e-mail address here
Title (tentative): Spectral Graph Learning: Algorithm and Application to Image Coding & Graph Convolutional Nets
Institute of Global Innovation Research, “LIFE SCIENCE" Yuichi Tanaka Team, Tokyo University of Agriculture and Technology
Excellent Leader Development for Super Smart Society by New Industry Creation and Diversity, Tokyo University of Agriculture and Technology
Workshop (1:30–2:20 PM, March 4, 2021)
We will have a workshop at 1:30-2:20 PM, March 4, 2021. We will invite Prof. Yu Tsao (Acadmia Sinica) as a lecturer. This workshop will be held by Zoom. Anyone can join for free. Please register your name and e-mail address here to join the event. If you have any questions, contact us (higashi-h [at] i.kyoto-u.ac.jp).
ワークショップを3月4日に開催します.Prof. Yu Tsao (Acadmia Sinica) の招待講演をオンラインで2021年3月4日午後1時30分〜午後2時20分で開催します.本講演会はZoomで開催されます.無料でどなたでも聴講できます.聴講するにはこちらから名前,メールアドレスをご登録ください.ご質問があれば higashi-h [at] i.kyoto-u.ac.jp までご連絡ください.
Please register your name and e-mail address here.
Title: Deep-learning-based Speech Enhancement with Its Application to Assistive Oral Communications Devices
Abstract: Speech enhancement (SE) serves as a key component in most speech-related applications. The goal of SE is to enhance the speech signals by reducing distortions caused by additive and convoluted noises in order to achieving improved human-human and human-machine communication efficacy. In the this talk, we will review the system architecture and fundamental theories of deep learning based SE approaches. Next, we will present more recent advances, including end-to-end and goal-driven based SE systems as well as the SE systems with improved architectures and feature extraction procedure. The reinforcement learning and generative adversarial network (GAN)-based SE methods will also be presented. Finally, we will discuss some applications based on the deep learning SE systems, including impaired speech transformation and noise reduction for assistive hearing and speaking devices.
Yu Tsao received the B.S. and M.S. degrees in electrical engineering from National Taiwan University, Taipei, Taiwan, in 1999 and 2001, respectively, and the Ph.D. degree in electrical and computer engineering from the Georgia Institute of Technology, Atlanta, GA, USA, in 2008. From 2009 to 2011, he was a Researcher with the National Institute of Information and Communications Technology, Tokyo, Japan, where he engaged in research and product development in automatic speech recognition for multilingual speech-to-speech translation. He is currently a Research Fellow (Professor) and Deputy Director with the Research Center for Information Technology Innovation, Academia Sinica, Taipei. His research interests include speech and speaker recognition, acoustic and language modeling, audio coding, and bio-signal processing. He is currently an Associate Editor for the IEEE/ACM Transactions on Audio, Speech, and Language Processing and IEEE Signal Processing Letters and a Distinguished Lecturer of APSIPA. He was the recipient of the Academia Sinica Career Development Award in 2017, the National Innovation Award in 2018, 2019, 2020, Future Tech Breakthrough Award 2019, and the Outstanding Elite Award, Chung Hwa Rotary Educational Foundation 2019–2020.
Technical Committee of Signal Processing (SIP), Institute of Electronics, Information and Communication Engineers (IEICE)
Technical Committee of Speech (SP), Institute of Electronics, Information and Communication Engineers (IEICE)
Technical Committee of Engineering acoustics (EA), Institute of Electronics, Information and Communication Engineers (IEICE)
Special Interest Group on Spoken Language Processing (SLP), Information Processing Society of Japan (IPSJ)
Online Kick-off Event and Seminars (9:50–12:00 AM, November 5, 2020)
The following seminars will be held at 9:50–12:00 AM, November 5, 2020. This workshop will be held by Zoom. Anyone can join for free. Please register your name and e-mail address in the link to join the event. If you have any questions, contact us (higashi-h [at] i.kyoto-u.ac.jp).
以下の講演会をオンラインで2020年11月5日午前9時50分〜12時で開催します.本講演会はZoomで開催されます.無料でどなたでも聴講できます.聴講するにはこちらから名前,メールアドレスをご登録ください.ご質問があれば higashi-h [at] i.kyoto-u.ac.jp までご連絡ください.
Please register your name and e-mail address here.
9:50–10:00 Opening remarks
Hitoshi Kiya (APSIPA President, Tokyo Metropolitan University) and Yoshinobu Kajikawa (APSIPA Vice-President, Kansai University)
10:00–11:00 Masahito Togami (LINE Corporation)
Title: Deep speech source separation considering spatial model
Abstract: Deep neural network (DNN) is a powerful modeling tool for speech source spectrum. Although the DNN is an important technique, we believe that combination of the DNN with signal modeling based approaches, e.g., spatial model, is quite important. The signal models based on solid scientific knowledge can be expected to compensate for performance degradation when there is little data for learning or when there is mismatch between the learning environment and the evaluation environment. In this presentation, I will present several modern speech source separation techniques followed by our deep speech source separation which is combined with spatial modeling.
Masahito Togami received B.E., M.E., and Ph.D. degrees in aerospace engineering from the University of Tokyo in 2000, 2002, and 2011, respectively. He started artificial intelligence research of Hitachi's communication robot, EMIEW1-3. He received several awards such as the Awaya Award in 2009, the Itakura Award in 2010 from the Acoustical Society of Japan (ASJ), and the 29th TELECOM System Technology Award from the Telecommunications Advancement Foundation in 2014. He is a member of the ASJ and the JSAI and a senior member of the IEEE. He is a Board Member of the JSAI. He is also a sponsorship Co-Chairs of APSIPA2021. He joined LINE Corporation in 2018. He is now the manager and a principal researcher of speech team in LINE Corporation. He wrote ``Speech source separation with Python'' in japanese this year.
11:00–12:00 Antonio Ortega (University of Southern California)
Title: DeepNNK: A polytope interpolation framework for graph-based neural network analysis
Abstract: Modern machine learning systems based on neural networks have shown great success in learning complex data patterns while being able to make good predictions on unseen data points. However, these methods are hindered by time consuming model selection and lack of predictive explainability, especially in the presence of adversarial examples. In this talk, we take a step towards better understanding of neural networks by introducing a local polytope interpolation method. The proposed Deep Non Negative Kernel regression (NNK) interpolation framework is non parametric, theoretically simple and geometrically intuitive. We present the NNK graph construction, demonstrate instance based explainability and develop a method to identify models with good generalization properties using leave one out estimation.
Antonio Ortega is a professor of Electrical and Computer Engineering at the University of Southern California (USC). He is a Fellow of the IEEE and EURASIP, and a member of ACM and APSIPA. He is the Editor-in-Chief of the IEEE Transactions of Signal and Information Processing over Networks and recently served as a member of the Board of Governors of the IEEE Signal Processing Society. He has received several paper awards, including the 2016 Signal Processing Magazine award. His recent research work is focusing on graph signal processing, machine learning, multimedia compression and wireless sensor networks.