Autonomous Intelligent Agents: Technology and Policy
Recent progress in the control of autonomous agents coupled with advances in artificial intelligence put us on the verge of profound changes in the ways in which we live, work and play; they will also push the boundaries in how cities are managed, how society polices itself and they will change how governments behave and are perceived. Our speakers and panel members will explore the technical opportunities and policy challenges in these areas considering, for example, how our attitudes and decision making may change as we both commission and interact with intelligent agents that are acting on our behalf and/or on the behalf of others.
These agents may be designing an individual’s wellness program and diet for the week. They may be driving children to school or delivering packages by air—all with a level of autonomy, decision making and, in many cases, an assumed ability to mitigate risk and/or appropriately improvise when encountering unique scenarios. The symposium is free and open to the public.
Tuesday, April 26, 2016
|9:15 a.m.–9:30 a.m.||NAE Members-Only Registration||Building 92 | Lobby|
|9:30 a.m.–10:15 a.m.||Self-Guided Tour||Building 92 | Visitor Center|
|10:30 a.m.–12:30 p.m.||NAE Members-Only Business Meeting||Building 92 | Origami Conference Room|
|12:30 p.m.–1:00 p.m.||Symposium Registration||Building 92 | Lobby|
|1:00 p.m.–1:05 p.m.||Symposium Kickoff, Dr. Daron Green, Managing Director, Microsoft Outreach||
Building 92 |
|1:05 p.m.–1:15 p.m.||Welcome from the NAE, Dr. C. D. Mote, Jr., President, NAE|
|1:20 p.m.–1:50 p.m.||Myths and Facts About AI, Dr. Oren Etzioni, Chief Executive Officer, Allen Institute for Artificial Intelligence|
|1:55 p.m.–2:25 p.m.||Law and Emergent Behavior, Ryan Calo, Assistant Professor of Law, University of Washington|
|2:30 p.m.–3:00 p.m.||Designing Increasingly Autonomous Vehicles to Be Safer than Humans, Dr. Linda Ng Boyle, Professor, Industrial and System Engineering, University of Washington|
|3:00 p.m.–3:15 p.m.||BREAK|
|3:15 p.m.–3:45 p.m.||AI, Autonomy, and Human-Computer Collaboration, Eric Horvitz, Technical Fellow and Managing Director, Microsoft Research|
|3:50 p.m.–4:20 p.m.||Benefits and Concerns of Automation in Commercial Aviation, Dr. Barbara Holder, Technology Fellow, Honeywell | Aerospace|
|4:25 p.m.–5:00 p.m.||Panel Discussion | Q&A (Linda Ng Boyle, Barbara Holder, Eric Horvitz)|
|5:00 p.m.–7:00 p.m.||Reception||Building 92 | Second Floor|
Linda Ng Boyle
Topic: Designing Increasingly Autonomous Vehicles to Be Safer than Humans
Linda Ng Boyle is professor and chair in the Department of Industrial & Systems Engineering at the University of Washington and holds a joint appointment in Civil & Environmental Engineering. She is also Director of the Human Factors and Statistical Modeling Lab, and an Associate Faculty with the Harborview Injury Prevention & Research Center. Her research interests center on understanding people’s behavior while driving vehicles and the impact on mishaps, injuries, and crashes. Injuries are the fourth leading cause of deaths in the US and those from motor vehicle crashes are preventable. In her work she has found that many innovations are designed to increase a person’s safety by simplifying demands and tasks in safety-critical situations. However, long-term use may also decrease their awareness, increase driver distractions, and degrade the ability to respond appropriately. Modeling this dynamic adaptive behavior is the core of many of her research studies. Dr. Boyle received her MS and PhD from the U. Washington, and is a recipient of an NSF Career Award. She serves as an associate editor for Accident Analysis & Prevention and is the Chair of the Transportation Research Board’s Committee on Statistical Methods.
Topic: Law and Emergent Behavior
Ryan Calo is an assistant professor at the University of Washington School of Law and an assistant professor (by courtesy) at the Information School. He is a faculty co-director (with Batya Friedman and Tadayoshi Kohno) of the University of Washington Tech Policy Lab, a unique, interdisciplinary research unit that spans the School of Law, Information School, and Department of Computer Science and Engineering. Professor Calo is a CoMotion Presidential Innovation Fellow for the class of 2015.
Professor Calo’s research on law and emerging technology appears or is forthcoming in leading law reviews (California Law Review, University of Chicago Law Review, Stanford Law Review Online, University of Pennsylvania Law Review Online) and technical publications (MIT Press, IEEE, Science, Artificial Intelligence), and is frequently referenced by the mainstream media (NPR, New York Times, Wall Street Journal). Professor Calo has also testified before the full Judiciary Committee of the United States Senate and spoken at the Aspen Ideas Festival and NPR’s Weekend in Washington. In 2014, he was named one of the most important people in robotics by Business Insider.
Professor Calo is an affiliate scholar at the Stanford Law School Center for Internet and Society (CIS), where he was a research fellow, and the Yale Law School Information Society Project (ISP). He serves on numerous advisory boards, including the University of California’s People and Robots Initiative, the Electronic Frontier Foundation (EFF), the Electronic Privacy Information Center (EPIC), Without My Consent, and the Future of Privacy Forum.
Professor Calo worked as an associate in the Washington, D.C. office of Covington & Burling LLP and clerked for the Honorable R. Guy Cole on the U.S. Court of Appeals for the Sixth Circuit. Prior to law school at the University of Michigan, Professor Calo investigated allegations of police misconduct in New York City.
Topic: Myths and Facts About AI
Dr. Oren Etzioni is Chief Executive Officer of the Allen Institute for Artificial Intelligence. He has been a Professor at the University of Washington’s Computer Science department since 1991, receiving several awards including GeekWire’s Hire of the Year (2014), Seattle’s Geek of the Year (2013), the Robert Engelmore Memorial Award (2007), the IJCAI Distinguished Paper Award (2005), AAAI Fellow (2003), and a National Young Investigator Award (1993). He was also the founder or co-founder of several companies including Farecast (sold to Microsoft in 2008) and Decide (sold to eBay in 2013), and the author of over 100 technical papers that have garnered over 25,000 citations. The goal of Oren’s research is to solve fundamental problems in AI, particularly the automatic learning of knowledge from text. Oren received his Ph.D. from Carnegie Mellon University in 1991, and his B.A. from Harvard in 1986.
Topic: Benefits and Concerns of Automation in Commercial Aviation
Barbara Holder received her M.S. and Ph.D. in Cognitive Science from the University of California, San Diego, specializing in human-centered systems design. She holds a B.A. in Social Ecology from the University of California Irvine, specializing in environmental design and public policy, and is a licensed pilot. After completing a Post-doctoral fellowship investigating how pilots acquire expertise in the use of the A320 flight management systems, Dr. Holder transitioned to Aviation Safety in Boeing Commercial Airplanes Group. While at Boeing she led numerous research projects and conducted ethnographic research at airlines around the world to investigate pilot performance in interaction with technology. She research resulted in the redesign of the normal and emergency checklists and pilot procedures, pilot and instructor training, electronic flight bag applications, new flight management concepts, and airplane systems interfaces for advanced flight decks. Dr. Holder also served as a subject matter expert on several industry committees and working groups including FAA aviation rulemaking committees for Loss of Control In-flight, Air Carrier Training, and Flight Path Management. She also serves on the International Federation of Airline Pilots Association Human Performance committee and on the editorial board for the Journal of Cognitive Engineering and Decision Making. Dr. Holder recently transitioned to Honeywell Aerospace as a Technology Fellow in Advanced Technology where she is responsible for Honeywell’s crew interface technology strategy. She is Chair of the Crew Interface Technology Council that defines the strategic direction of new crew interface concepts. She continues to focus her research on enhancing human performance in complex socio-technical systems and in applying cognitive ethnography to explore new interaction concepts for advanced flight decks.
Topic: AI, Autonomy, and Human-Computer Collaboration
Eric Horvitz is technical fellow and director of the Microsoft Research lab at Redmond. His research centers on probabilistic and decision-theoretic representations and inference procedures. His contributions include efforts in healthcare, transportation, information retrieval, and human-computer interaction. He has been elected fellow of AAAI, AAAS, the American Academy of Arts and Sciences, and the National Academy of Engineering, and he has been inducted into the CHI Academy. He received the Feigenbaum Prize for contributions to AI. He has served as president of the AAAI, chair of the AAAS Section on Information, Computing, and Communications, and on advisory committees for the National Institutes of Health, National Science Foundation’s CISE Directorate, PCAST NITRD Working Group, Computing Community Consortium (CCC), Naval Research Advisory Committee (NRAC), and the Information and Science Study Group (ISAT) of DARPA. More information can be found here.
Myths and Facts about AI
Speaker: Dr. Oren Etzioni, Chief Executive Officer, Allen Institute for Artificial Intelligence
AI’s recent success has led to excess. We see headlines like : Artificial Intelligence is Coming, and it Could Wipe Us Out if We’re Not Careful, Professor Warns. While some successes are real (for example, AlphaGo’s amazing Go playing), many challenges remain. My talk will put AlphaGo (and related learning systems) in context, and attempt to debunk some of the popular myths about AI. I will conclude by talking about AI2’s mission of AI for the Common Good—as illustrated by our AI-based scientific search engine: Semantic Scholar.
Law and Emergent Behavior
Speaker: Ryan Calo, Assistant Professor of Law, University of Washington
Drawing from contemporary and fictional examples, this talk by a law professor and leading authority on robotics law and policy will cover some of the ways civil and criminal law may react to emergent behavior on the part of intelligent agents. Audience questions and participation are particularly welcome.
Designing increasingly autonomous vehicles to be safer than humans
Speaker: Dr. Linda Ng Boyle, Professor Industrial and System Engineering, University of Washington
Vehicles are becoming increasingly autonomous but are not as yet all fully driverless. Autonomous technology is being introduced in phases. This requires human drivers to continually adapt to their changing role inside the car. The human’s ability to decide when to take back control of the primary driving task is further complicated given system imperfections. As drivers adapt to increasing levels of automation, they may change their driving behavior in ways that can compromise and even negate the intended system benefits. To design vehicles that account for limitations of both the human and the system, it is essential to model such behavioral adaptation. This presentation will focus on the challenges and opportunities in this modeling endeavor and their safety implications for all road users.
AI, Autonomy, and Human-Computer Collaboration
Speaker: Eric Horvitz, Technical Fellow and Managing Director, Microsoft Research
I will discuss opportunities ahead for bringing together the complementary abilities of human and machine intelligence. I will start by reflecting about the rich design space for human-machine teaming along a spectrum of autonomy. I will present several mechanisms that enable effective human-computer collaboration and review examples of the use of machine intelligence to augment human decision making and to work with people to jointly solve problems. I will then turn to automated perception, learning, and inference for high-stakes decisions in the open world, rife with complexity and unforeseen situations. I will discuss critical research needs, including developing methods that endow systems with the ability to infer their competencies in different settings and to understand and to plan when and how to engage people to assist when needed.
Benefits and Concerns of Automation in Commercial Aviation
Speaker: Dr. Barbara Holder, Technology Fellow, Honeywell | Aerospace
The global commercial air transport system has seen record low accident rates for the past ten years. Although automation has advanced the safety and efficiency of the system it remains a contributing factor in accidents and incidents due to clumsy interaction with pilots. A logical conclusion might be to automate the pilots out of the interaction. The issue with this direction is there will always be humans in the system so our focus should be on developing technologies that enable humans to effectively collaborate with automation.