Keynotes


Unlocking Autonomous Mobility

Vinay Shet, Director of Product Management, Lyft, San Francisco, California
Wednesday, November 6

Abstract: Autonomous cars are expected to dramatically redefine the future of transportation. When fully realized, this technology promises to unlock a myriad societal, environmental, and economic benefits. From a technical standpoint, autonomous cars pose significant challenges. An autonomous car needs to perceive the world around it, it needs to maintain a precise geographic context of its location, it needs to predict the future state of all dynamic agents, and finally, it needs to plan its own trajectory to progress to its destination.
In this talk, I will discuss Lyft's mission to build the world's best transportation network and how autonomous cars play a key role in realizing it. I will then present our approach to building autonomous cars breaking down the technical components, digging deeper into perception, prediction, planning, and HD Mapping. I will also discuss our efforts to democratize access to autonomous technology to the broader academic community through the release of open datasets and hosting academic competitions. With this, I hope to share insights into the motivations, challenges, and technologies for autonomous cars from the perspective of an advanced industrial program.

Biography: Vinay Shet is a Director of Product Management at Lyft Level 5 leading the Mapping, Data, and Knowledge efforts for Lyft’s self-driving car initiative. Before Lyft, Vinay was at Google as the product lead for Google Maps’ Road Network and Navigation data. At Google, he led the team that pioneered the use of machine learning to create maps data from street level and aerial imagery. Before that, he was the Product Manager for Google Street View as well as reCAPTCHA, launching the “I’m not a robot” CAPTCHA. Vinay has a Ph.D. in Computer Science focusing on Computer Vision from the University of Maryland, College Park and was a Scientist at Siemens Corporate Research, focused on Machine Learning and Computer Vision, for several years post-Ph.D. He has published over 20 papers and has over 25 issued/pending patents. He has served as a reviewer/program committee member of multiple computer vision journals, conferences, and workshops such as CVPR, ECCV, ICCV, AVSS, etc.


Visual Spatial Analytics and Trusted Information for Effective Decision Making

David S. Ebert, Silicon Valley Professor of Electrical and Computer Engineering, Purdue University, West Lafayette, Indiana
Thursday, November 7

Abstract: Information, not just data, is key to today's global challenges. To solve these challenges requires not only advancing geospatial and big data analytics but requires new analysis and decision-making environments that enable reliable decisions from trustable, understandable information that go beyond current approaches to machine learning and artificial intelligence. These environments are successful when they effectively couple human decision making with advanced, guided spatial analytics in human-computer collaborative discourse and decision making (HCCD). Our HCCD approach builds upon visual analytics, natural scale templates, traceable information, human-guided analytics, and explainable and interactive machine learning, focusing on empowering the decision-maker through interactive visual spatial analytic environments where non-digital human expertise and experience can be combined with state-of-the-art and transparent analytical techniques. When we combine this approach with real-world application-driven research, not only does the pace of scientific innovation accelerate, but impactful change occurs. I'll describe how we have applied these techniques to challenges in sustainability, security, resiliency, public safety, and disaster management.

Biography: David Ebert is the Silicon Valley Professor of Electrical and Computer Engineering at Purdue University, a Fellow of the IEEE, director of the Center for Education and Research in Information Assurance and Security (CERIAS), and director of the Visual Analytics for Command Control and Interoperability Center (VACCINE), the Visualization Science team of the Department of Homeland Security's Command Control and Interoperability Emeritus Center of Excellence. Ebert performs research in visual analytics, human-computer teaming, advanced predictive analytics, volume rendering, illustrative visualization, and procedural abstraction of complex, massive data. He is the recipient of the 2017 IEEE Computer Society vgTC Technical Achievement Award for seminal contributions in visual analytics. He has been very active in the visualization community, serving as Editor in Chief of IEEE Transactions on Visualization and Computer Graphics, serving as IEEE Computer Society Vice President and the IEEE Computer Society's VP of Publications, and successfully managing a large program of external funding to develop more effective methods for visually communicating information.