Title: Efficient Multimodal Diffusion-Policies
Bio: Rudolf Lioutikov is a tenure track professor for machine learning and robotics at the Karlsruhe Institute of Technology, Germany. He started the Intuitive Robots Lab in June 2021 after being accepted into the Emmy-Noether Programme by the German Research Foundation. The Lab develops new methods to facilitate human-robot interaction and collaboration. Previously Rudolf was an Assistant Professor of Practice at the University of Texas at Austin. Rudolf was awarded his Ph.D. with distinction by the Technical University of Darmstadt in 2018. His dissertation on the "Imitation Learning Pipeline" was nominated as a finalist for the Georges Giralt PhD Award by the European Robotics Federation.
Title: Robots Learning Through Interactions
Bio: Jens Kober is an associate professor at the TU Delft, Netherlands. He worked as a postdoctoral scholar jointly at the CoR-Lab, Bielefeld University, Germany and at the Honda Research Institute Europe, Germany. He graduated in 2012 with a PhD Degree in Engineering from TU Darmstadt and the MPI for Intelligent Systems. For his research he received the annually awarded Georges Giralt PhD Award for the best PhD thesis in robotics in Europe, the 2018 IEEE RAS Early Academic Career Award, the 2022 RSS Early Career Award, and has received an ERC Starting grant. His research interests include motor skill learning, (deep) reinforcement learning, imitation learning, interactive learning, and machine learning for control.
Title: Zero-Shot, One-Shot, and Few-Shot Robot Learning
Abstract: In recent years, behavioural cloning has been shown to work very well for robot manipulation when a very large number of demonstrations are available, and several companies now appear to be adopting this strategy with a view towards developing commercial robots. But what if the company selling you this future robot did not train the robot on the specific tasks you actually need? You would not want to spend hours providing demonstrations for each of these tasks. As such, rather than relying on scaling up data collection, my team and I have instead been studying very efficient imitation learning with just one or a few demonstrations per task, and zero-shot reasoning with foundation models without needing any demonstrations at all. In this talk, I will present a number of such methods for vision-based robot manipulation, including two of our ICRA 2024 papers.
Bio: Edward Johns is the Director of the Robot Learning Lab at Imperial College London, where he is also a Senior Lecturer (Associate Professor). He received a BA and MEng from Cambridge University, and a PhD from Imperial College. He was then a post-doc at UCL, before returning to Imperial College as a founding member of the Dyson Robotics Lab. In 2017, he was awarded a Royal Academy of Engineering Research Fellowship, and then in 2018 he was appointed as a Lecturer and founded the Robot Learning Lab. In this lab, Ed and his team are currently developing methods for robots to efficiently learn everyday tasks in everyday environments.
Title: Language-Conditioned Robot Learning
Title: Redundant Kinematics Learning for Tensegrity Manipulator
Bio: Shuhei Ikemoto received his Ph.D. degree in engineering from Osaka University in March 2010. Since April 2010, he has been an assistant professor at the Graduate School of Information Science and Technology, The Institute for Academic Initiatives, and the Graduate School of Engineering Science at Osaka University. He is now an associate professor at the Graduate School of Life Science and Systems Engineering, Kyushu Institute of Technology since April 2019. His research interests include biologically inspired robots and algorithms, soft robotics, and physical human-robot interaction.
Title: Robotics in Action: Learning to Navigate, Manipulate, and Cooperate
Bio: ASAKO KANEZAKI received the B.S., M.S. and Ph.D. degrees in information science and technology from The University of Tokyo, in 2008, 2010, and 2013, respectively. In 2010, she was a Visiting Researcher with the Intelligent Autonomous Systems Group, Technische Universität München. From 2013 to 2016, she was an Assistant Professor with The University of Tokyo. She was with the National Institute of Advanced Industrial Science and Technology (AIST), from 2016 to 2020. Since 2020, she has been an Associate Professor with the Tokyo Institute of Technology. Her current research interests include object detection, 3D shape recognition, and robot applications, such as semantic mapping and visual navigation.
Title: Honda Avatar Robot and Learning for Manipulation
Bio: Akinobu Hayashi is an Assistant Chief Engineer at Frontier Robotics, Innovative Research Excellence at Honda R&D Co., Ltd. He received his master's degree in engineering from the University of Tokyo in 2009. After joining Honda R&D in Japan in April 2009, he spent several years at Honda Research Institute Europe GmbH in Germany, from 2013 to 2019. His research focuses on planning under uncertainty, probabilistic reasoning, imitation learning, and reinforcement learning. Akinobu is currently working on developing algorithms that facilitate robust, scalable, and contact-rich manipulation.
Title: Reinforcement Learning in the Age of Gigantic Datasets (or: RL do we still need it?)
Bio: Markus Wulfmeier is a researcher in machine learning and robotics at Google DeepMind with a focus on fundamental and applied research on reinforcement and imitation learning. His work aims at efficiently scalable algorithms applicable across a variety of real-world applications including robotic locomotion, manipulation, navigation, and communication. Markus was a postdoctoral research scientist at the Oxford Robotics Institute and a member of Oxford University’s New College where he completed his PhD. Over the years, he has held visiting scholar positions with UC Berkeley, MIT, and ETH.