Biography: Ming C. Lin is currently Distinguished University Professor and Elizabeth Stevinson Iribe Chair of Computer Science at the University of Maryland College Park and John R. & Louise S. Parker Distinguished Professor Emerita of Computer Science at the University of North Carolina (UNC), Chapel Hill. She was also an Honorary Visiting Chair Professor at Tsinghua University. She obtained her B.S., M.S., and Ph.D. in Electrical Engineering and Computer Science from the University of California, Berkeley. She received several honors and awards, including the NSF Young Faculty Career Award, UNC Hettleman Award for Scholarly Achievements, Beverly W. Long Distinguished Professorship, IEEE VGTC Virtual Reality Technical Achievement Award, Washington Academy Distinguished Career Award, and many best paper awards at international conferences. She is a Fellow of National Academy of Inventors, ACM, IEEE, and Eurographics, ACM SIGGRAPH Academy and IEEE VR Academy.
Her research interests include computational robotics, haptics, physically-based modeling, virtual reality, sound rendering, and geometric computing. She has (co-)authored more than 300 refereed publications in these areas and co-edited/authored four books. She has served on hundreds of program committees of leading conferences and co-chaired dozens of international conferences and workshops. She is currently a member of Computing Research Association-Women (CRA-W) Board of Directors, and was Chair of IEEE Computer Society (CS) Fellows Selection Committee, Chair of IEEE CS Computer Pioneer Award, and Founding Chair of ACM SIGGRAPH Outstanding Doctoral Dissertation Award. She is a former member of IEEE CS Board of Governors, a former Editor-in-Chief of IEEE Transactions on Visualization and Computer Graphics (2011-2014), a former Chair of IEEE CS Transactions Operations Committee, and a member of several editorial boards.
Title: From Learning-based Traffic Reconstruction to Autonomous Driving
Abstract: Rapid urbanization and increasing traffic have led to digitalization of modern cities and automation of transportation means. As new technologies like VR systems and self-driving cars emerge, there is an increasing demand to incorporate realistic traffic flows into virtualized cities. In this talk, we first present a novel method for learning-based traffic animation and visualization using GPS data. This approach reconstruct city-scale traffic using statistical learning on GPS data and metamodel-based simulation optimization for dynamic data completion in areas of insufficient data coverage. We also propose a novel differentiable hybrid traffic simulator, which simulates trafficusing a hybrid model of both macroscopic and microscopic models and can be directly integrated into a neural network for traffic control and flow optimization, being the first differentiable traffic simulator for macroscopic and hybrid models that can compute gradients for traffic states across time steps and inhomogeneous lanes. Next, we present a unified collision avoidance algorithm for the navigation of arbitrary agents, from pedestrians to various types of robots, including vehicles, in a unifying framework using different nonlinear motion extrapolations of motion to support agent dynamics. We then describe a learning-based, multi-level control policy for autonomous vehicles by analyzing simulated accident data and using our collision avoidance algorithm, data annotation, and parameterized traffic & vehicle simulation. We further introduce a simple yet effective framework for improving the robustness of learning algorithm against (input) image corruptions for autonomous driving, due to both internal (e.g., sensor noises and hardware abnormalities) and external factors (e.g., lighting, weather, visibility, and other environmental effects). We conclude by suggesting possible future directions.
Biography: Yi Ma is a Professor at the Department of Electrical Engineering and Computer Sciences at the University of California, Berkeley. His research interests include computer vision, high-dimensional data analysis, and intelligent systems. Yi received his Bachelor’s degrees in Automation and Applied Mathematics from Tsinghua University in 1995, two Masters degrees in EECS and Mathematics in 1997, and a PhD degree in EECS from UC Berkeley in 2000. He has been on the faculty of UIUC ECE from 2000 to 2011, the principal researcher and manager of the Visual Computing group of Microsoft Research Asia from 2009 to 2014, and the Executive Dean of the School of Information Science and Technology of ShanghaiTech University from 2014 to 2017. He then joined the faculty of UC Berkeley EECS in 2018. He has published about 60 journal papers, 120 conference papers, and three textbooks in computer vision, generalized principal component analysis, and high-dimensional data analysis. He received the NSF Career award in 2004 and the ONR Young Investigator award in 2005. He also received the David Marr prize in computer vision from ICCV 1999 and best paper awards from ECCV 2004 and ACCV 2009. He has served as the Program Chair for ICCV 2013 and the General Chair for ICCV 2015. He is a Fellow of IEEE, ACM, and SIAM.
Title: Compressive Closed-Loop Transcription from the Principles of Parsimony and Self-Consistency
Abstract: Ten years into the revival of deep networks and artificial intelligence, we propose a theoretical framework that sheds light on understanding deep networks within a bigger picture of intelligence in general. We introduce two fundamental principles, Parsimony and Self-consistency, that address two fundamental questions regarding Intelligence: what to learn and how to learn, respectively. We argue that these two principles can be realized in entirely measurable and computable ways. We demonstrate this with modeling visual data with an important family of structures and models, namely an arrangement of independent subspaces, known as a linear discriminative representation (LDR). The two principles naturally lead to an effective and efficient computational framework, known as a compressive closed-loop transcription, that unifies and explains the evolution of modern deep networks and modern practices of artificial intelligence. In this framework, we will see how fundamental elements in information theory, control theory, game theory, and optimization are naturally integrated in such a closed-loop system all as necessary ingredients. Such a system is the key to autonomously learn multi-class structures from real-world data and represent them in compact and structured forms. This work reveals that a central mathematical and computational problem for intelligence is how to learn and transform multiple nonlinear submanifolds in high-dimensional spaces to linear subspaces. While we mainly use modeling of visual data as an example, we believe that two principles are the cornerstones for the emergence of intelligence, artificial or natural, and the compressive closed-loop transcription is a universal learning engine that serve as the basic learning units for all autonomous intelligent systems, including the brain. Related papers can be found at: https://arxiv.org/abs/2207.04630 and https://www.mdpi.com/1099-4300/24/4/456/htm.
Biography: Xinde Li earned his Ph.D. in Control Theory and Control Engineering, from Department of Control Science and Engineering, Huazhong University of Science and Technology (HUST), Wuhan, China, in 2007. Afterwards, he joined School of Automation, Southeast University, Nanjing, China, where he is currently a professor. He was an Academician of Russian Academy of Natural Sciences elected in 2020 and a cover person of Scientific Chinese in the same year. He was a vice director of Intelligent Robot Committee of Chinese Association for Artificial Intelligence from 2017, a vice director of Intelligent Products and Industry Working Committee of Chinese Association for Artificial Intelligence from 2019. His research interests include Artificial Intelligence, Intelligent Robot, Machine Perception and Understanding, and human-robot interaction, etc. He has undertaken many national key projects, i.e. National 863 key project, JKW key project, etc. and has published more than 80 high quality papers and 2 books, and owns 17 national invention patents. He also won many prizes, i.e. international contribution prize, Scientific and Technological Progress Award in CAA, etc.
Title : Occasion Cognition in Coexisting -cooperative -cognitiove Robot
Abstract: With the advances of artificial intelligence, robots have been widely used in many service sectors and close to human life. However, constructing a harmonious and natural human-robot interaction environment puts forward high requirements for humanoid performance of robot. In this presentation, we first summarize the natural person and place in the robot interaction environment as the occasion, proposing the concept of robot occasion cognition. Then, according to the challenges of robot cognition for human and place in the occasion, I will report the recent researches about human identification and emotion recognition, place perception and understanding, in which we explore a new paradigm of robot occasion cognition towards various people and local conditions. Finally, I will give some prospects for the future progress of this topic.
Biography: Dr. Jee-Hwan Ryu is an Associate Professor in the Department of Civil and Environmental Engineering at Korea Advanced Institute of Science and Technology (KAIST). He received the B.S. degree in mechanical engineering from Inha University, South Korea, in 1995, and the M.S. and Ph.D. degrees in mechanical engineering from KAIST, South Korea, in 1997 and 2002, respectively. From 2002 to 2003, he worked as a post-doc researcher in the department of electrical engineering at the University of Washington, and at the similar time, he was also affiliated with the institute of robotics and mechatronics in DLR as a visiting scientist. He joined KAIST in 2019 as an associate professor. Prior to that, he was a professor in the department of mechanical engineering at KOREATECH (2005-2019), and a research professor in the department of electrical engineering at KAIST (2003-2005). His research interests include haptics, telerobotics, exoskeletons, and autonomous vehicles. He has received several awards including IEEE Most Active Technical Committee Award as a Co-chair of TC Haptics in 2015, Best poster award in 2010 IEEE Haptic Symposium, Outstanding reviewer award in 2020 IEEE ICRA, and Best paper award in 2021 IEEE Robotics and Automation Magazine. He has served as an Associate Editor in IEEE Transactions on Haptics, and since 2020, he has been serving as a Senior Editor in IEEE Robotics and Automation Letters. From 2017 to 2019, he has served as an Associate Editor-in-chief in World Haptics Conference, and Editor-in-chief in the 2021 World Haptics Conference. He was involved in many international conference organizations, and especially, he has been served as a general chair of AsiaHaptics2018.
Title: How Stiff or Light we can Reach: Time-domain Passivity Approach for Stable and Transparent Haptic Interaction
Abstract: The addition of haptic capability dramatically increases the immersiveness of human- robot interaction in AR/VR or teleoperation. That is because the sense of touch conveys rich and detailed information about virtual or remote environments. However, it has been challenging to provide immersive feelings of touch due to the limited range of impedance that a haptic device can display without any stability issue. In this presentation, we will be discussing how to realize stable and immersive human-robot haptic interaction, in particular from the aspect of tight haptic coupling between human and virtual/remote environment. Several state-of-the-art control methods, such as Time Domain Passivity Approach, Successive Stiffness Increment, Successive Force Augmentation method will be introduced, which have been developed for increasing the impedance range of both impedance type and admittance type haptic interfaces for the interaction with virtual objects and remote environments. A stable bilateral teleoperation method to overcome time varying communication delay will be introduced as well with several implementation examples with DLR space telerobotic systems.