Want to distribute CoRL handouts? Printable version is here
CoRL 2017 LaTex style files can be downloaded here.


The 1st Annual Conference on Robot Learning (CoRL 2017) is soliciting contribution at the intersection of robotics and machine learning. CoRL will be established as a selective, top-tier venue for robot learning research, covering a broad range of topics spanning robotics and ML, including both theory and practice:
  • Reinforcement learning
  • Model learning and control
  • State estimation, mapping, and computer vision
  • Multimodal perception and sensor fusion
  • Learning-based human robot interaction, natural language instruction processing
  • Applications in manipulation, mobility, driving, flight, and other areas of robotics
  • Bio-inspired learning and control
Robotics, autonomous perception, and control are undergoing a machine learning revolution. Now is the time to dedicate a venue that will combine fundamental advances in machine learning with an empirical exploration of robotics applications and theory. CoRL will be organized as a focused three-day meeting, with invited keynote talks and contributor presentations. We invite cutting-edge research papers to be submitted to CoRL 2017. Accepted papers will published via the JMLR Workshop & Conference Proceedings series and will be presented either as posters or oral presentations in the plenary session. Our aim is to make CoRL the top venue for robot learning research.

Mountain View, California, USA

Key dates
Paper submission deadline: June 28, 2017 (both archival and workshop track)
Paper acceptance notification: September 1, 2017
Conference dates: November 13-15, 2017



Papers may be submitted through CMT, via the link at the top of this page. Submissions are due June 28, 2017 at 5:00 pm PT. Submissions may be submitted to one of two tracks, as described below. Both tracks will use the same single-column Latex template, which may be found at the top of this page. CoRL is double-blind, which means all papers must be anonymized. Authors are permitted to post their paper e.g. on arxiv, but the submission itself must not contain any overt indication of the authors’ identities. The two tracks are:

  • Archival track: The archival track will consist of papers up to eight pages in length (plus up to two additional pages of references). Authors will have the option to submit a supplementary file containing further details, which the reviewers may decide to consult, as well as a supplementary video. All supplementary materials will be submitted through CMT as a single zip file. Archival track papers will be reviewed by three reviewers. Accepted papers will appear in the JMLR Workshop & Conference Proceedings as officially published CoRL papers. All accepted papers will be provided with a poster presentation slot, and about 20-30 papers will be presented as oral presentations in the main conference. While it is difficult to judge how many papers will be submitted or accepted, since this is the first year for CoRL, we expect a large fraction of accepted papers to receive oral presentation slots.

  • Non-archival track: The non-archival track will consist of short 4-page papers (plus up to one additional page of references). The policy on supplementary materials will be the same as above. Accepted papers will not appear in any official archival format. For accepted non-archival papers, authors may request the CoRL website to link to a full-length version (e.g., on arxiv). Non-archival track papers will be presented in a poster session. A very small number of non-archival track papers may receive oral presentation slots. The non-archival track is intended for both “workshop-style” papers reporting late breaking results and for more mature work that the authors are submitting to another archival venue. Reviewers will be asked to consider both late breaking but less polished results and more polished papers. Non-archival track papers are subject to a different dual submission policy -- see below.

Dual Submission Policy

  • Archival track: We will not accept papers that are identical or substantially similar to papers that have previously been published or accepted for publication in an archival venue, nor papers submitted in parallel to other conferences. Archival venues include conferences and journals with formally published proceedings, but do not include non-archival workshops. Submission is permitted for papers that have previously appeared only as a technical report.

  • Non-archival track: The non-archival track will accept papers that have been submitted or accepted for publication at other venues, but have not yet been presented (e.g. at conferences); we will accept short versions of journal articles that appeared in print within the past 12 months; we will not accept conferences papers that were previously accepted and already presented. Papers that are rejected from the archival track may be referred to the non-archival track at the reviewers’ discretion. Please do not submit the same paper to both the archival and non-archival track.

Reviewing Criteria

Submissions will be evaluated based significant novel results. Results may be either theoretical or empirical. Results will be judged on the degree to which they have been objectively established and/or their potential for scientific and technological impact, as well as their relevance to robotic learning. We take a broad view of robotic learning. Papers with both experimental and theoretical results relevant to robotic learning are welcome. Our intent is to make CoRL into a selective top-tier conference on robotic learning.

Journal of Machine Learning Research: proceedings publication