Workshop on Self-Supervised Robot Learning

Robotics: Science and Systems
Virtual Workshop

News

• 2020-07-13: The workshop has ended. We will upload the recordings in the next days. Thank you for attending!
• 2020-07-13: The best paper award goes to Manderson et al., for their paper "Self-Supervised, Goal-Conditioned Policies for Navigation in Unstructured Environments". 
• 2020-07-09: Updated website with spotlight talks of contributed papers.
• 2020-07-06: Updated website with the live stream link and accepted papers.
• 2020-07-03: Updated website with schedule, registration details, and presentation format.
• 2020-06-29: List of accepted papers are now online.
• 2020-06-28: All authors have been notified of the paper decisions.
• 2020-06-24: We will be awarding a Best Paper Award in cooperation with Google Research.
• 2020-03-15: UPDATE CONCERNING COVID-19: The RSS 2020 SSRL organizing committee hope you are safe and well. Due to the pandemic and the uncertainty regarding travel to the US, the RSS 2020 SSRL workshop will take place VIRTUALLY on the original date of July 13th.

Overview

Self-supervised learning is an exciting research direction that aims to learn representations from the data itself without explicit and potentially even manual supervision. One of the major benefits of self-supervised learning is the ability to scale to large amounts of unlabelled data in a lifelong learning manner and to improve performance by reducing the effect of dataset bias. Recent development in self-supervised learning has resulted in achieving comparable or better performance than fully-supervised models. However, many of these methods are developed in domain-specific communities such as robotics, computer vision or reinforcement learning. The aim of this workshop is to bring together researchers from different communities to discuss opportunities, challenges and explore new directions.

Recorded talks will be available on our YouTube channel. Live talks and panel discussions will be streamed through RSS PheedLoop and YouTube Live.

Topics

The focus topics of our workshop include, but are not restricted to:

• Self-supervised learning for robotics, robot vision, reinforcement learning….
• Self-supervised domain adaptation
• Meta-learning of self-supervised tasks
• Large-scale self-supervised learning
• Learning of generalizable pretext-tasks
• Loss functions for self-supervised learning
• Learning from auxiliary/multiple tasks
• Multimodal and cross-modal learning

Invited Speakers

 

Call for Contributions

We encourage participants to submit their research in the form of a single PDF. Submissions may be up to 4 pages in length, including figures, excluding references and any supplementary material. Please use the RSS conference template. Accepted papers will be presented in a poster session and selected awards papers as spotlight talks. All submitted contributions will go through a single blind review process. The contributed papers will be made available on the workshop’s website and selected papers will be invited for a special issue of a major robotics journal.

Submission Website: https://easychair.org/conferences/?conf=ssrl20

LaTeX Template: https://roboticsconference.org/docs/paper-template-latex.tar.gz

In order to make acceptance decisions early, we request interested researchers to submit a single page extended abstract by 19th April as an expression of interest. The authors would then have time 2nd July to include new results and submit the full 4 page paper. We also welcome submissions that have already been accepted to RSS or other major conferences and journal papers that have not been discussed in a conference.

Please note that we will be awarding a Best Paper Award in cooperation with Google Research.

Registration

• Attendees and speakers are requested to register for the main RSS 2020 conference which gives you access to all the workshops
• Registered attendees will be able to participate in the workshop via Zoom which can be accessed from the RSS PheedLoop virtual conference platform
• Registered attendees will be able to discuss with the speakers during the panel discussions or use the Q&As via Slack which can be accessed from the RSS PheedLoop Engage tab
• All the workshop talks will be streamed live via YouTube for those who are not registered and the archived talks will be made available here after the workshop concludes

Invited Talks and Panel Discussions

• Invited talks will be live and will take 18 min each
• Invited speakers will be participating in the subsequent panel discussions
• Selected questions will be taken from the previously collected Google form, as well as questions asked via Slack which can be accessed from RSS PheedLoop, and YouTube live (moderated)

Contributed Papers

• Each contributed paper will be introduced as a pre-recorded 5 min spotlight talk
• The talks will be made online on 9th July 2020 here
• The attendees are requested to watch the pre-recorded talks before the workshop, 13th July 2020
• The papers will be discussed with a Q&A during the contributed paper discussion sessions

Organizers

 

For further information please contact us at rss20-ssrl(at)googlegroups.com