Kyushu University invites four professors from Peking University and holds Public Seminars on computational intelligence, speech and hearing, perceptual psychology, and image processing. No registration and registration fee are requested.
We are welcome not only students and faculty members but also anyone from outside of Kyushu University,
日時： 12月5日（月） / Dec. 5 (Mon) 16:30～17:30
場所： 九州大学伊都キャンパス ウエスト２号館５階システム情報第２会議室（547）/ ISEE Meeting Room 2 (Room 547), 5F, West Zone Building No.2, Ito Campus
講師： 譚営教授 / Prof. Ying Tan （北京大学 信息科学技術学院 / School of EE and CS, Peking University）
題目： "Research Advances in Fireworks Algorithm and Its Applications"
日時： 12月5日（月） / Dec. 5 (Mon) 13:00～15:00
場所： 九州大学大橋キャンパス 3号館6階601 / Room 601, 6F, Building No.3, Ohashi Campus
講師： 羅定生副教授 / Prof. Dingsheng Luo （北京大学 信息科学技術学院 / School of EE and CS, Peking University）
題目： "Bridging gaps for Human-Human and Human-Machine: A Perspective from Speech and Hearing Research"
日時： 12月17日（土） / Dec. 17 (Sat) 17:00～18:00
場所： 九州大学大橋キャンパス3号館322教室 / Room 322, 2F, Building No.3, Ohashi Campus
講師： 陳立翰副教授 / Prof. Lihan Chen （北京大学 心理・認知科学学院 / School of Psychological and Cognitive Science）
題目： "Temporal perceptual grouping and transfer in a multisensory context"
日時： 12月20日（火） / Dec. 20 (Tue) 15:00～16:00
場所： 九州大学伊都キャンパス ウエスト2号館3F システム情報科学府第5+6講義室 / ISEE No.5+6 room, 3F, West Zone Building No.2, Ito Campus
講師： 張超副教授 / Prof. Chao Zhang （北京大学 信息科学技術学院 / School of EE and CS, Peking University）
題目： "Subspace Recovery Based Feature Learning"
交通とキャンパス地図 / Access and Campus Map
問合せ (contact)： 高木英行 （九州大学芸術工学研究院・教授）
Advances in Fireworks Algorithm and Its Applications
譚営（TAN Ying）教授 （北京大学 信息科学技術学院）
12月5日（月） / Dec. 5 (Mon) 16:30～17:30
ISEE Meeting Room 2 (Room 547), 5F, West Zone Building No.2, Ito Campus
協賛： IEEE SMC Japan Chapter，IEEE 福岡Section， IEEE Computational Intelligence Society Japan Chapter，進化計算学会，計測自動制御学会 (九州支部，コンピューテーショナル・インテリジェンス部会)
Recently, inspired from the collective behaviors of many swarm-based creatures in nature or social phenomena, swarm intelligence (SI) has been received attention and studied extensively, gradually becomes a class of efficiently intelligent optimization methods. Inspired by fireworks explosion at night, the fireworks algorithm (FWA) was developed in 2010. Since then, several improvements and some applications were proposed to improve the efficiency of FWA. In this talk, the fireworks algorithm is first described in detail and reviewed, then several effective improved fireworks algorithms are highlighted individually. By changing the ways of calculating numbers and amplitudes of sparks in fireworks’ explosion, the improved FWA algorithms become more reasonable and explainable. In addition, the multi-objective fireworks algorithm and the graphic processing unit (GPU) based FWA are also briefly presented, particularly the GPU-based FWA is able to speed up the optimization process considerably. Extensive experiments on IEEE-CEC’s benchmark functions demonstrate that the improved fireworks algorithms significantly increase the accuracy of found solutions, yet decrease the running time dramatically. Finally, some applications of FWA are briefly described, while its shortcomings and future research directions are identified.
| Ying Tan is a full professor and PhD advisor at the
School of Electronics Engineering and Computer Science of Peking
University, and director of Computational Intelligence Laboratory at
Peking University (PKU). He received his BEng, MS, and PhD from
Southeast Univ., in 1985, 1988, and 1997, respectively. He is the
inventor of Fireworks Algorithm (FWA).
He serves as the Editor-in-Chief of International Journal of Computational Intelligence and Pattern Recognition (IJCIPR), the Associate Editor of IEEE Transaction on Cybernetics (Cyb), the Associate Editor of IEEE Transaction on Neural Networks and Learning Systems (NNLS), International Journal of Artificial Intelligence (IJAI), International Journal of Swarm Intelligence Research (IJSIR), etc. He also served as an Editor of Springer’s Lecture Notes on Computer Science (LNCS) for more than 20 volumes, and Guest Editors of several referred Journals, including Information Science, Soft Computing, Neurocomputing, IJAI, IJSIR, B&B, CJ, IEEE/ACM Transactions on Computational Biology and Bioinformatics (IEEE/ACM TCBB). He received the 2nd-Class Natural Science Award of China in 2009 and a number of academic prizes in his fields. He is a senior member of IEEE. He is the founder and chair of the ICSI International Conference series. He was one of joint general chairs of 1st&2nd BRICS CCI, program committee co-chair of WCCI 2014, etc.
His research interests include computational intelligence, swarm intelligence, data mining, pattern recognition, intelligent information processing for information security. He has published more than 260 papers in refereed journals and conferences in these areas, and authored/co-authored 10 books and 12 chapters in book, and received 4 invention patents.
Bridging gaps for Human-Human and Human-Machine:
A Perspective from Speech and Hearing Research
羅定生（LUO Dingsheng）副教授 （北京大学 信息科学技術学院）
12月5日（月） / Dec. 5 (Mon) 13:00～15:00
Room 601, 6F, Building No.3, Ohashi Campus
Human society becomes more and more rich, colorful and convenient in the light of the rapid development of the science and technology. Still, we suffer from gaps either between human and human due to disabilities derived from diseases or aging, or between human and machine facing challenges from the complex and versatile real world. As it has been turned out, researches on speech and hearing, including perception mechanism, computational modeling, processing technologies as well as efficient algorithms etc., do bring significant improvements in either helping the rehabilitation of hearing impaired people, or promoting the more effective and convenient interfaces for intelligent systems. To bridging such gaps which are multidisciplinary and full of challenges, with the establishment of the speech and hearing research center at Peking University, interdisciplinary researches such as mechanisms of auditory perception and speech production, hearing aids and cochlear implant, auditory scene analysis and acoustic field reconstruction, speech recognition and synthesis, language understanding and dialogue management, learning techniques as well as embodied cognition based humanoids etc., are being carried out. Along with the discussions of the motivation, challenges and research architectures, recent research progresses will also be discussed in this talk. Although those gaps may not be completely eliminated in the near future, we believe they are on the way of gradually shrinking.
| Dingsheng Luo, Associate Professor, Ph.D, the Deputy
Director of the Department of Machine Intelligence, Peking University.
He received his Ph.D. degree from Peking University, China, in 2003.
Then, he joined the Peking University till now, and became an Associate
Professor in 2005.
Since 2003, he is with the Key Laboratory of Machine Perception (Ministry of Education), the Speech and Hearing Research Center, the Department of Machine Intelligence, School of the EECS, Peking University, China.
He has/had undertaken more than 20 research projects funded by Chinese Government, is the author of more than 60 research papers and 5 inventions, awarded Best Paper Award of CNNC in 2002, Zhengda Award of Peking University in 2008, First Prize Award of IJCAI-13 Robot Competition Technical Skills Category in 2013 and Best Paper Finalist Award of IEEE ICIA in 2016.
He is an IEEE member, a member of IEEE-RAS Technical Committee on Humanoid Robotics, a member of the Climbing and Walking Robots Association, a senior member of Chinese Association for Artificial Intelligence. He serves as Deputy Secretary General of the Chinese Association for Artificial Intelligence Educational Committee. He also serves as reviewer of several Domestic or international journals, and as session chairs and PC members of several international conferences, such as AAAI.
His research interests includes: Artificial Intelligence, Speech and Language Processing, Machine Learning, Robotics, Humanoid Robot, Human-Machine/Robot Interaction etc.
Temporal Perceptual Grouping and Transfer in a Multisensory Context
陳立翰（CHEN Lihan）副教授 （北京大学 心理・認知科学学院）
12月17日（土） / Dec. 17 (Sat) 17:00～18:00
Room 322, 2F, Building No.3, Ohashi Campus
Multisensory processing within seconds is pivotal for many perceptual and cognitive tasks. Here we adopted the research paradigms of temporal ventriloquism, priming and illusory (auditory) gap transfer to address the guiding principles of intra-modal and cross-modal temporal perceptual grouping. We used methods of psychophysics, fMRI and MEG to investigate the (ensemble) coding of temporal events and the underpinning neural signatures. The accumulated evidence has also shown that our brain takes a centralized temporal representation and the benefits of temporal training of events in one sensory modality could be quickly transferred to another modality.
Keywords: Temporal; Perceptual grouping; Ternus display; psychophysics; MEG; fMRI
| Lihan Chen is an Associate Professor at School of
Psychological and Cognitive Sciences, Peking University. He obtained his
Bachelor degree in Biomedical Engineering from Zhejiang University in
1999 and Master degree in General Psychology from Zhejiang University in
2005. He got his PhD degree in Experimental Psychology at Ludwig
Maximilians University Munich in 2010, working with Dr. Zhuanghua Shi
and Dr. Hermann J. Müller.
He was a postdoc fellow between 2009 and 2011 and was an Assistant Professor during 2011-2015 at Department of Psychology, Peking University. His major research interests include multisensory time perception,cross-modal correspondence and tactile perception.
Recovery Based Feature Learning
張超（ZHANG Chao）副教授 （北京大学 信息科学技術学院）
12月20日（火） / Dec. 20 (Tue) 15:00～16:00
九州大学伊都キャンパス ウエスト2号館3F システム情報科学府第5+6講義室
ISEE No.5+6 lecture room, 3F, West Zone Building No.2, Ito Campus
共催： 九州大学，IEEE Computer Society Fukuoka Chapter
協賛： IEEE Systems, Man and Cybernetics Society Japan Chapter
Subspace recovery based feature learning methods is a kind of feature learning method which is proposed recently. This kinds of methods aim at characterizing the relationship between data samples and then learning more discriminative feature by recovering the inherent subspace structure of data. Actually, this kinds of methods have been successfully applied to machine learning and computer vision problems, such as face recognition, object classification, background modeling, and visual saliency. In this talk I will describe recent research in our lab on subspace recovery based feature learning that attempts to combines feature learning with classification, so that the regulated classification error is minimized. In this way, the extracted features are more discriminative for the recognition tasks.
|Chao Zhang is an Associate Professor at the School of Electronics Engineering and Computer Science of Peking University. He received the Ph.D. degree in electrical engineering from Beijing Jiaotong University, Beijing, China, in 1995. He was a Post-Doctoral Research Fellow with the National Laboratory on Machine Perception, Peking University, Beijing, from 1995 to 1997. He has been an associate professor with the Key Laboratory of Machine Perception(MOE), School of Electronics Engineering and Computer Science, Peking University since 1997. His current research interests include image processing, statistical pattern recognition, and visual recognition. He has published more than 60 papers in refereed journals and conferences in these areas, and received 6 national invention patents.|