Featured Talks

2. Invited Speakers

Vladimir Cherkassky
Dept. of Electrical & Computer Engineering
University of Minnesota
Minneapolis MN 55455
email cherk001@umn.edu
http://www.ece.umn.edu/users/cherkass/predictive_learning/ 

Technical Bio: Vladimir Cherkassky is Professor of Electrical and Computer Engineering at the University of Minnesota, Twin Cities. He received MS in Operations Research from Moscow Aviation Institute in 1976 and PhD in Electrical and Computer Engineering from the University of Texas at Austin in 1985. He has worked on theory and applications of statistical learning since late 1980’s and he has co-authored the monograph Learning from Data, now in its second edition. He is also the author of a new textbook Predictive Learning www.VCtextbook.com
He has served on editorial boards of IEEE Transactions on Neural Networks (TNN), Neural Networks, Natural Computing, and Neural Processing Letters. He was a Guest Editor of the IEEE TNN Special Issue on VC Learning Theory and Its Applications published in September 1999. Dr. Cherkassky was organizer and Director of NATO Advanced Study Institute (ASI) From Statistics to Neural Networks: Theory and Pattern Recognition Applications held in France in 1993. He received the IBM Faculty Partnership Award in 1996 and 1997 for his work on learning methods for data mining. In 2007, he became Fellow of IEEE for ‘contributions and leadership in statistical learning’. In 2008, he received the A. Richard Newton Breakthrough Award from Microsoft Research for ‘development of new methodologies for predictive learning’.

Title: Methodological Aspects of VC-theory

Abstract: Vapnik-Chervonenkis theory (VC-theory), aka Statistical Learning Theory, provides mathematical framework for predictive data-analytic methods developed in machine learning, statistics, data mining, signal processing, bioinformatics etc. The VC-theory was developed in the late 1970’s in the former USSR. However, it became widely known only in mid-1990’s after introduction of SVMs. Current interest in VC-theory Theory can be measured by 31 million hits on Google search; likewise search on Support Vector Machines yields 6.9 million hits. Even though VC-theory is widely known as a mathematical theory, its methodological contributions are not well known or appreciated. This talk focuses on VC-theoretical methodology for estimating data-analytic predictive models. In particular, I will discuss important differences between knowledge discovery in classical science and modern data-analytic knowledge discovery. The classical framework encourages a popular view that knowledge can be discovered by applying readily available statistical/ machine learning software to the growing volumes of data, expressed as a view known as Big Data: more_data→ more_knowledge. An opposite view is that scientific inquiry starts with asking intelligent questions. That is, Science starts from problems, and not from observations (K. Popper). For data-analytic modeling, this view emphasizes proper formalization of application domain requirements and selecting proper type of inductive inference. To this end, VC-theoretical methodology clearly separates the problem setting from a learning method or algorithm. This talk will describe several non-standard learning settings (developed in VC-theory) and contrast them to standard inductive setting adopted in most machine learning and statistical algorithms. Various methodological issues will be illustrated using real-life application examples.

Anders Sandberg

Bio: Anders Sandberg is a researcher, science debater, futurist, transhumanist and author. He holds a Ph.D. in computational neuroscience from Stockholm University, and is currently a James Martin Research Fellow at the Future of Humanity Institute at Oxford University. Sandberg's research centres on societal and ethical issues surrounding human enhancement and new technology, as well as on assessing the capabilities and underlying science of future technologies. His recent contributions include work on cognitive enhancement (methods, impacts, and policy analysis); a technical roadmap on whole brain emulation; on neuroethics; and on global catastrophic risks, particularly on the question of how to take into account the subjective uncertainty in risk estimates of low-likelihood, high-consequence risk. He is well known as a commentator and participant in the public debate about human enhancement internationally, as well as for his academic publications in neuroscience, ethics, and future studies. He is co-founder of and writer for the think tank Eudoxa, and is a co-founder of the Orion's Arm collaborative world building project. Between 1996 and 2000 he was Chairman of the Swedish Transhumanist Association. He was also the scientific producer for the neuroscience exhibition "Se Hjärnan!" ("Behold the Brain!"), organized by Swedish Travelling Exhibitions, the Swedish Research Council and the Knowledge Foundation, that toured Sweden in 2005–2006. In 2007 he was a postdoctoral research fellow at the Uehiro Centre for Practical Ethics at Oxford University, working on the EU-funded ENHANCE project on the ethics of human enhancement.

Title: Ethics and large-scale neural simulations: when do we need to start caring for networks, rather than about them?

Abstract: <TBA> 

Cesare Alippi

Bio: CESARE ALIPPI received the degree in electronic engineering cum laude in 1990 and the PhD in 1995 from Politecnico di Milano, Italy. Currently, he is a Full Professor of information processing systems with the Politecnico di Milano. He has been a visiting researcher at UCL (UK), MIT (USA), ESPCI (F), CASIA (RC), A*STAR (SIN). Alippi is an IEEE Fellow, Distinguished lecturer of the IEEE CIS, Member of the Board of Governors of INNS, Vice-President education of IEEE CIS, Associate editor (AE) of the IEEE Computational Intelligence Magazine, past AE of the IEEE-Trans. Instrumentation and Measurements, IEEE-Trans. Neural Networks, and member and chair of other IEEE committees. In 2004 he received the IEEE Instrumentation and Measurement Society Young Engineer Award; in 2013 he received the IBM Faculty Award. Among the others, Alippi was General chair of the International Joint Conference on Neural Networks (IJCNN) in 2012, Program chair in 2014, Co-Chair in 2011. He was General chair of the IEEE Symposium Series on Computational Intelligence 2014. Current research activity addresses adaptation and learning in non-stationary environments and Intelligence for embedded systems. Alippi holds 5 patents, has published in 2014 a monograph with Springer on “Intelligence for embedded systems” and (co)-authored more than 200 papers in international journals and conference proceedings. Home Page: http://home.dei.polimi.it/alippi/

Title: Intelligence for Cyber-Physical and Embedded Systems

Abstract: The emergence of non-trivial embedded sensor units and cyber-physical systems has made possible the design and implementation of sophisticated applications where large amounts of real-time data are collected to constitute a big data picture as time passes. Acquired data are then processed at local, cluster-of-units or server level to take the appropriate actions or make the most suitable decision. Within this framework, intelligence mechanisms play a key role to provide systems with advanced functionalities. Intelligent mechanisms are needed to optimally harvest and manage the residual energy, identify possible faults affecting some sensors, reduce the communication bandwidth, solve the compromise between accuracy and computational complexity, guarantee appropriate performances within an evolving, time invariant environment. The talk will show how the use of intelligence can boost the next generation of embedded and cyber-physical-based applications, generation whose footprint is already around us.