Arun Ross, Michigan State University, United States
Title: Recent Progress in Biometrics
Ludmila Kuncheva, School of Computer Science, Bangor University, United Kingdom
Title: Getting Lost in the Wealth of Classifier Ensembles?
Fabio Roli, Università degli Studi di Cagliari, Italy
Title: Adversarial Pattern Recognition
Tanja Schultz, Cognitive Systems Lab (CSL), University of Bremen, Germany
Title: Biosignal-based Cognitive Systems and Applications
Recent Progress in Biometrics
Michigan State University
Arun Ross is an Associate Professor in the Department of Computer Science and Engineering at Michigan State University (MSU) and the Director of the i-PRoBe Lab. Prior to joining MSU in 2013, he was a faculty member at West Virginia University (WVU) from 2003 to 2012. He also served as the Assistant Site Director of the NSF Center for Identification Technology and Research (CITeR) between 2010 and 2012.
Arun received the B.E. (Hons.) degree in Computer Science from the Birla Institute of Technology and Science, Pilani, India, and the M.S. and Ph.D. degrees in Computer Science and Engineering from Michigan State University. He is the coauthor of the textbook “Introduction to Biometrics” and the monograph “Handbook of Multibiometrics,” and the co-editor of “Handbook of Biometrics”. He is a recipient of the IAPR JK Aggarwal Prize, the IAPR Young Biometrics Investigator Award (YBIA), the NSF CAREER Award, and was an invited speaker at the Frontiers of Science Symposium organized by the National Academy of Sciences in November 2006. He is also a recipient of the 2005 Biennial Pattern Recognition Journal Best Paper Award and the Five Year Highly Cited BTAS 2009 Paper Award.
Arun served as a panelist at a counter-terrorism event that was organized by the United Nations Counter-Terrorism Committee (CTC) at the UN Headquarters in May 2013. He was an Associate Editor of IEEE Transactions on Information Forensics and Security (2009 – 2013), and IEEE Transactions on Image Processing (2008 – 2013). He currently serves as Area Editor of the Computer Vision and Image Understanding Journal, Associate Editor of the Image and Vision Computing Journal, and Chair of the IAPR TC4 on Biometrics.
Biometrics is the science of recognizing individuals based on their physical and behavioral attributes such as fingerprints, face, iris, voice and signature. The past decade has witnessed tremendous progress in this field, including the deployment of biometric solutions in diverse applications such as border security, national ID cards, amusement parks, access control, and smartphones. Despite these advancements, biometric systems have to contend with a number of challenges related to data quality, spoof attacks, and personal privacy. This talk will highlight some of the recent progress made in the field of biometrics; present our lab’s work on heterogeneous biometrics, fingerprint spoof detection, and biometric data privacy; and discuss some of the challenges that have to be solved in order to promote the widespread use of this technology.
Getting Lost in the Wealth of Classifier Ensembles?
School of Computer Science, Bangor University
Ludmila Kuncheva studied electrical engineering at the Technical University, Sofia, Bulgaria (1977-1982), where she obtained her MSc in 1982. She took a postgraduate course in applied mathematics, 1982-1984, followed by a PhD, received in 1987 from the Bulgarian Academy of Sciences. Until 1997 she worked at the Central Laboratory of Biomedical Engineering, Bulgarian Academy of Sciences, as a research associate and senior research associate. She worked at the ELITE laboratory (Aachen, Germany, 1993, European Mobility Grant), Imperial College (London, 1995-1996, Royal Society Fellowship) and University of West Florida (Pensacola, Florida, 1996-1997, NSF grant). Ludmila Kuncheva joined Bangor University, in 1997 as a lecturer at the School of Mathematics. She is currently a professor at the School of Computer Science at Bangor University. Professor Ludmila Kuncheva is working in the area of pattern recognition and machine learning. She has published over 150 research papers and two books:- "Fuzzy Classifier Design, Springer, 2000.", and "Combining Pattern Classifiers. Methods and Algorithms, Wiley, 2004". She won of the best paper award for 2006 in IEEE Transactions on Fuzzy Systems and the Sage best Transaction paper award for 2003 across IEEE Transactions on Systems, Man and Cybernetics, A, B and C. Ludmila Kunchevahas served as an AE for IEEE Transactions on Fuzzy Systems and for IEEE Transactions on Pattern Analysis and Machine Intelligence. She has given invited talks and lectures within the UK and abroad:- Spain, France, Italy, China, USA, The Netherlands, Bulgaria, Greece and Sweden.
Classifier ensembles have proven their worth for solving challenging problems of modern-day pattern recognition. It is easy to get lost in the massive volume of relevant literature, the quickly growing number of new methods and algorithms being proposed, and the wide spread of the ensemble research into numerous application areas. To be able to benefit from this wealth of knowledge, we need, metaphorically speaking, beacons, anchor points, and a well-organised warehouse. At the current stage, we need tools for assessment, benchmarking, comparison and structuring, more than we need new ensemble methods. In an attempt to address these issues, this talk will discuss some of the “anchor points” such as ensemble diversity and combination strategies, and will give a bibliometric perspective on the development of the area. Finally, we will indulge in a brief speculation of what the future may hold for classifier ensembles.
Adversarial Pattern Recognition
Università degli Studi di Cagliari
Fabio Roli is professor of computer engineering and director of the PRA Lab (http://pralab.diee.unica.it). His research over the past twenty years addressed the design of pattern recognition systems in the context of real applications. He played a leading role for the research field of multiple classifier systems. He is a prolific author with a value of h-index of 43 according to Google Scholar. He is Fellow of the IEEE and Fellow of the IAPR.
Pattern recognition systems are currently used in several applications, like biometric recognition, spam filtering, and intrusion detection in computer networks, which differ from the traditional ones. The difference lies in the fact that in these applications an intelligent, adaptive adversary can actively manipulate patterns with the aim of making the classifier ineffective, namely, with the aim of evading it. This kind of problem has been recently named adversarial classification, and is the subject of an emerging research field in the machine learning and pattern recognition communities.
In this talk, I introduce the fundamentals of adversarial classification from the perspective of a designer of pattern recognition systems, and illustrate the concepts of adversary-aware classifier, security evaluation, and defence countermeasures, with examples from security applications like spam filtering and biometric recognition.
Biosignal-based Cognitive Systems and Applications
Cognitive Systems Lab (CSL), University of Bremen
Tanja Schultz received her doctoral and diploma degree in Informatics from University of Karlsruhe, Germany in 2000 and 1995 respectively and successfully passed the German state examination for teachers of Mathematics, Sports, and Educational Science from Heidelberg University, in 1990.
She joined Carnegie Mellon University in Pittsburgh, Pennsylvania in 2000 and up to now holds a position as Research Professor at the Language Technologies Institute. From 2007 to 2015 she was a Full Professor at the Department of Informatics of the Karlsruhe Institute of Technology (KIT) in Germany before she accepted an offer from the University of Bremen in April 2015. Since 2007, Tanja Schultz directs the Cognitive Systems Lab, where her research activities focus on human-machine communication with a particular emphasis on multilingual speech processing and human-machine interfaces. Together with her team, she investigates the processing, recognition and interpretation of biosignals, i.e. human signals such as speech, motion, muscle and brain activities to enable human interaction with machines in an intuitive and efficient way.
Tanja Schultz received several awards for her work, such as the FZI award for an outstanding Ph.D. thesis (2001), the Allen Newell Medal for Research Excellence from Carnegie Mellon (2002), the ISCA best journal award for her publication on language independent acoustic modeling (2002) and on Silent Speech Interfaces (2015), the Plux Wireless award (2011) for the development of Airwriting, the Alcatel-Lucent Research Award for Technical Communication (2012), the Google Research Award and the Otto-Haxel Award (2013), as well as several best paper awards. She is the author of more than 280 articles published in books, journals, and proceedings, and is regularly invited as panelist and keynote speaker. Tanja Schultz serves as a member for numerous conference committees, as Associate Editor of IEEE Transactions (2002-2004), as an Associate Editor for ACM TALLIP (since 2013), as editorial board member of Speech Communication (since 2001), and currently serves as elected president of the International Speech Communication Association ISCA.
Biosignals are autonomous signals produced by humans measured in physical quantities. In the context of human-computer interaction, human modalities like speech, gestures or motion, i.e. muscle and brain activity at large, can be captured by non-invasive body-worn sensors. The processing and interpretation of the resulting biosignals offer an inside perspective on human physical and mental activities, intentions, and needs and thus complement the traditional way of observing human interaction from the outside.
As recent years have seen major advances in sensor and device technologies, including new monitoring methods, miniaturized sensors, as well as integrated, mobile and ubiquitous devices, the time is right to use the full range of biosignals for cognitive systems and applications. In my talk I will present ongoing research at the Cognitive Systems Lab (CSL), where we explore human-centered cognitive systems to improve human-machine interaction as well as machine-mediated human communication. Several applications will be described such as Silent Speech Interfaces that rely on articulatory muscle movement captured by electromyography to recognize and synthesize silently produced speech, as well as brain-computer interfaces that capture brain activity by electrocorticography to recognize speech and by electroencephalograhy to determine users' mental states, such as task activity, cognitive workload and attention. We hope that our research will lead to a new generation of cognitive systems, which are completely aware of the users' needs and provide an intuitive, efficient, robust, and adaptive input mechanism to interaction and communication.