Conference Keynote Speakers






Prof. Jan "Honza" Černocký,
 Faculty of Information Technology, Brno University of Technology, Czech Republic

"Making speech recognition work on low-resource languages - Brno Babel experience" Read more...less

Bio: http://www.fit.vutbr.cz/~cernocky
(Ing. [MS] 1993 Brno University of Technology (BUT); [PhD] 1998 Universite Paris XI and BUT) was with the Institute of Radio-electronics, BUT (Faculty of Electrical Engineering and Computer Science) as assistant professor from 1997. He was a post-doc in Hynek Hermansky’s OGI group in Portland, Oregon, in 2001. Since February 2002, he is with the Faculty of Information Technology (FIT), BUT as Associate Professor (Doc.) and since 2008, he serves as the Head of the Department of Computer Graphics and Multimedia. He has founded the BUT Speech@FIT research group in 1997 and serves as its executive director.

He has been involved with several European projects: SPEECHDAT-E (4th FP, technical coordination), SpeeCon, Multimodal meeting manager (M4, both 5th FP), Augmented Multiparty interaction (AMI, 6th FP), Augmented Multiparty Interaction with Distance Access (AMIDA, 6th FP), CareTaker (6th FP) and A-PiMod (7th FP). He was BUT’s principal investigator in EC-sponsored MOBIO project (7th FP) and is BUT’s PI in the Horizon 2020 BISON project. He has also coordinated several projects supported by Czech Ministries of Industry and Interior, Technology Agency of the Czech Republic (TACR) and he supervises projects funded by U.S.
Government (EOARD, DARPA, IARPA).

His research interests include signal processing and speech data mining.
He is author or co-author of more than 50 papers in journals and at conferences. He regularly serves as reviewer for major speech conferences and journals. He is on the scientific board of FIT, scientific board of Text-Speech-Dialogue conference, Odyssey and ASRU 2017 committees. He served on the board of Czechoslovak section of IEEE, IEEE SPS Conference Board, and on the IEEE SLTC technical committee. He served as co-chair of ICASSP 2011 in Prague and co-chair of IEEE ASRU 2013 in Olomouc.

As faculty member of FIT BUT, Dr. Cernocky is active in teaching, he is responsible for signal- and speech-processing courses in all levels of studies (bachelor, master, doctoral). He is Senior Member of IEEE and member of ISCA.




Satellite Event | Unsupervised Machine Learning
Seminar Keynote Speaker






Prof. Shai Ben-David,
 School of Computer Science University of Waterloo

"How far are we from having a satisfactory theory of clustering?" Read more...less

Abstract: Unsupervised learning is widely recognized as one of the most important challenges facing machine learning nowadays.However, unlike supervised learning, our current theoretical understanding of those tasks,and in particular of clustering, is very rudimentary. Although hundreds of clustering papers are being published every year, there is hardly any work reasoning about clustering independently of any particular algorithm, objective function, or generative data model.

My talk will focus on such clustering research.  I will discuss two aspects in which theory could play a significant role in guiding the use of clustering tools. The first is model selection - how should a user pick an appropriate clustering tool for a given clustering problem, and how should the parameters of such an algorithmic tool be tuned? In contrast with other common computational tasks, in clustering, different algorithms often yield drastically different outcomes. Therefore, the choice of a clustering algorithm may play a crucial role in the usefulness of an output clustering solution. However, there currently exist no methodical guidance for clustering tool selection for a given clustering task. I will describe some recent proposals aiming to address this crucial lacuna.

The second aspect of clustering that I will address is the complexity of computing a cost minimizing clustering (given some clustering objective function). Once a clustering model(or objective) has been picked, the task becomes an optimization problem.

While most of the clustering objective optimization problems are computationally infeasible, they are being carried out routinely in practice.This theory-practice gap has attracted significant research attention recently. I will describe some of the theoretical attempts to address this gap and discuss how close do they bring us to a satisfactory understanding of the computational resources needed for achieving good clustering solutions.

Bio: I grew up in Jerusalem. I studied physics, mathematics and psychology at the Hebrew University, receiving my PhD in pure math under the supervision of Saharon Shelah and Menachem Magidor. In I 1987 joined the CS Department at the Technion. I held visiting faculty positions at the Australian National University in Canberra (1997-8) and at Cornell University (2001-2004). In August 2004 I moved to Canada as a professor at the School of Computer Science at the University of Waterloo.

I have served as an editor in a couple of CS theory journals, as a Program Committee chair, and member of the steering committees for COLT and ALT (the major ML theory conferences) and several times as area chair for NIPS and ICML. 

My research interests span a wide spectrum of topics in the foundations of computer science and its applications, with a particular emphasis on statistical and computational machine learning. The common thread throughout my research is aiming to provide mathematical formulation and understanding of real world problems. In particular, I have been looking at popular machine learning and data mining paradigms that seem to lack clear theoretical justification (such as unsupervised learning, clustering, transfer learning, semi-supervised learning and crowd sourcing).

Silver Sponsor

      

Professional Association Sponsor

Afeka Tel Aviv Academic College of EngineeringACLP - Afeka Center for Language Processing