International Conference on Robots and Systems (IROS)

→ Link to the Online Meeting.

Dear fans of robots with interest in machine learning,

We are happy to announce our virtual live workshop on “New Horizons for Robot Learning” as part of the program of the International Conference on Intelligent Robots and Systems (IROS) 2020. With my fellow colleagues Elmar Rückert and Arne Wahrburg we have put together an interesting virtual live session and will discuss with some of the hottest names in the area. Get ready to meet and discuss with Pieter Abbeel, Dileep George, Sergey Levine, Jan Peters, Freek Stulp, Marc Toussaint, Patrick van der Smagt, and Georg von Wichert.

The session is scheduled for October 29^th from 16:00 – 19:30 h CET (08:00 – 11:30 h PDST) and we look forward for your participation. Registration for the IROS Conference is free, so do not miss the chance to learn and discuss about the future of robot learning with us.

So long! Elmar, Arne & Kime

List of Speakers

Workshop Schedule

Time (CET)*	Time (PDST)*	Names	Theme
16:00 – 16:10	08:00 – 08:10	Arne, Elmar, Kim	Welcome & Agenda
16:10 – 16:30	08:10 – 08:30	Jan Peters	Robot Learning Lessons from Motor Skill Learning
16:30 – 16:50	08:30 – 08:50	Freek Stulp	Common-sense exploration for Robotic Learning
16:50 – 17:10	08:50 – 09:10	Marc Toussaint	Learning to do as planned. Slides
17:10 – 17:30	09:10 – 09:30	Georg v. Wichert	Combining Machine Learning and first principle models
17:30 – 17:40	09:30 – 09:40	BREAK
17:40 – 18:00	09:40 – 10:00	Patrick van der Smagt	Learning models in models. Details to the talk at https://argmax.ai
18:00 – 18:20	10:00 – 10:20	Dileep George	Data-efficient learning for robotics
18:20 – 18:40	10:20 – 10:40	Sergey Levine	Data-driven Reinforcement Learning for Robotics
18:40 – 19:00	10:40 – 11:00	Pieter Abbeel	Robot learning for the real world
19:00 – 19:30	11:00 – 11:30	Open panel	What is the future of Robot Learning and what will it bring for society, industry etc.?
19:30 – 20:00	11:30 – 12:00	Virtual Apero	Get together with Music and chatting (among the organizers and speakers)

*Europe will switch to from DST to „winter time“ already on Oct 25, while the US will switch no earlier than Nov 1st => time diff for the workshop will be -8 h (instead of the normal -9 h)

Workshop Summary

Robot learning combines the challenges of understanding, modeling and applying dynamical systems with task learning from rewards, through human robot interaction or from intrinsic motivation signals. While outstanding results using machine and deep learning have been generated in robot learning in the last years, current challenges in industrial applications are underrepresented. The goal of this workshop is to go beyond discussing potential industrial applications like in related past workshops.

The core elements of this event are industrial specific challenges, novel machine and deep learning approaches and unsolved complex problems. For that we will bring together leading experts from academia and industry to discuss current and future challenges in the application of learning mechanisms in industry. We try to have a problem-based look on robot learning: which methods would really help to advance robot learning applications and what are the demands from industry in the first place. Spiced with a discussion between various industrial as well as academic leaders this is unmatched by any classical conference session.

Organizers

Dr. Kim Daniel Listmann, Head of ABB Future Labs Switzerland, kim.listmann@ch.abb.com, +41 79 325 9946

ABB Switzerland Ltd, Segelhofstrasse 1K, 5405 Baden-Dättwil, Switzerland.

Prof. Dr. Elmar Rueckert, Professor at the Institute for Robotics and Cognitive Systems, rueckert@rob.uniluebeck.de, +49 451 310 15 209

University of Luebeck, Ratzeburger Allee 160, 23562 Luebeck, Germany.

Dr. Arne Wahrburg, Senior Principal Scientist, ABB Corporate Research Germany, arne.wahrburg@de.abb.com, +49 6203 716166

ABB AG, Wallstadter Str. 59, 68526 Ladenburg, Germany.