Keynote Abstracts:

Eric Horvitz:

Sensors. Predictions, and Decisions

Abstract: I will discuss directions with harnessing machine learning and inference to make predictions and to guide actions from sensed data, drawing examples from work on transportation, healthcare, and interactive systems.  I will start with efforts to field predictive models that forecast flows of traffic in greater city regions. Moving from the ground to the air, I will present work on fusing data from aircraft to make inferences about atmospheric conditions and the promise of using these results to enhance air transport. I will then turn to work on fielding predictive models in clinical medicine, including efforts to extend models by sensing and reasoning about space and time. Moving beyond data captured by hospitals, I will describe the promise of transforming anonymized behavioral data drawn from web services into large-scale sensor networks for public health, including recent efforts to identify adverse effects of medications and to understand illness in populations. Finally, I will discuss leveraging sensing and reasoning to develop physically-situated interactive systems that can collaborate with people in a fluid manner.

Bio: Eric Horvitz is a distinguished scientist and managing director at the Microsoft Research Lab at Redmond, Washington.  His interests span theoretical and practical challenges with machine learning, inference, and decision making.  He has been elected a fellow of AAAI, ACM, AAAS, and the National Academy of Engineering, and has been inducted into the CHI Academy.  He has served as president of the AAAI and chair of the AAAS Section on Information, Computing, and Communications, and on the advisory committees for the NSF’s Directorate for Computer & Information Science & Engineering (CISE), the Computing Community Consortium (CCC), and the DARPA Information and Science Study Group (ISAT).

John Lygeros:

Population Control

Abstract: Many large scale systems involve the interaction of a number of agents with loosely coupled dynamics and decisions. Examples include transportation systems, consumer demand response in electricity grids, emergency evacuation of buildings, and even education. In all these cases agents locally optimize their decisions, but their eventual well being depends on the decisions of all other agents. For such systems it is typically impractical to impose a centralized control structure for a number of reasons, including computational and communication limitations and privacy concerns. Instead one can consider providing suitable information to the agents and imposing an appropriate penalty/reward scheme to steer the overall population using macroscopic commands, so that it exhibits a desirable macroscopic behaviour. In this talk we will discuss how such control structures can be developed by adopting a mean field control perspective, based on convergence results in operator theory. The discussion will be motivated by applications of population control to consumer electricity demand response schemes.

Bio: John Lygeros completed a B.Eng. degree in electrical engineering in 1990 and an M.Sc. degree in Systems Control in 1991, both at Imperial College of Science Technology and Medicine, London, U.K.. In 1996 he obtained a Ph.D. degree from the Electrical Engineering and Computer Sciences Department, University of California, Berkeley. During the period 1996-2000 he held a series of research appointments at the National Automated Highway Systems Consortium, Berkeley, the Laboratory for Computer Science, M.I.T., and the Electrical Engineering and Computer Sciences Department at U.C. Berkeley. Between 2000 and 2003 he was a University Lecturer at the Department of Engineering, University of Cambridge, U.K., and a Fellow of Churchill College. Between 2003 and 2006 he was an Assistant Professor at the Department of Electrical and Computer Engineering, University of Patras, Greece. In July 2006 he joined the Automatic Control Laboratory at ETH Zurich, first as an Associate Professor, and since January 2010 as a Full Professor; he is currently serving as the Head of the laboratory. His research interests include modelling, analysis, and control of hierarchical, hybrid, and stochastic systems, with applications to biochemical networks, automated highway systems, air traffic management, power grids and camera networks. John Lygeros is a Fellow of the IEEE, and a member of the IET and the Technical Chamber of Greece.

Elizabeth Croft:

Up Close and Personal with Human-Robot Collaboration  

Abstract: Advances in robot control, sensing and intelligence are rapidly expanding the potential for close-proximity human-robot collaborative work. In many different contexts, from manufacturing assembly to home care settings, a robot’s potential strength, precision and process knowledge can productively complement human perception, dexterity and intelligence to produce a highly coupled, coactive, human-robot team. Such interactions, however, require task-appropriate communication cues that allow each party to quickly share intentions and expectations around the task.  These basic communication cues allow dyads, human-human or human-robot, to successfully and robustly pass objects, share spaces, avoid collisions and take turns – some of the basic building blocks of good, safe, and friendly collaboration regardless of one’s humanity.  In this talk we will discuss approaches to identifying, characterizing, and implementing communicative cues and validating their impact in human-robot interaction scenarios.    

Bio: Elizabeth A. Croft, Ph.D., P.Eng., Fellow Engineers Canada, Fellow American Society of Mechanical Engineers, is a Professor of Mechanical Engineering and Associate Dean, Education and Professional Development for the Faculty of Applied Science at UBC.  She holds the NSERC Chair for Women in Science and Engineering, BC-Yukon at UBC and leads the WWEST program for women in engineering, science and technology.  As director of the Collaborative Advanced Robotics and Intelligent Systems (CARIS) Laboratory at UBC, Dr. Croft’s research investigates how robotic systems can behave, and be perceived to behave, in a safe, predictable, and helpful manner, and how people interact with and understand robotic systems. Applications of this work range from manufacturing assembly to healthcare and assistive technology. Elizabeth received a Peter Wall Early Career Scholar award in 2001, the Association of Professional Engineers and Geoscientists (BC) Professional Service Award in 2005, the Award for the Support of Women in the Engineering Profession, Canadian Council of Professional Engineers in 2006, an NSERC Accelerator award in 2007, a YWCA Women of Distinction Award in 2013, and one of WXN’s 2014 top 100 most powerful women in Canada.