Call for papers

• We solicit submissions for two separate tracks within the workshop. All submissions should follow the ICLR2023 formatting guidelines (please use the template at https://polybox.ethz.ch/index.php/s/06eBpVZECfayWnk) and should be submitted at https://openreview.net/group?id=ICLR.cc/2023/Workshop/Trustworthy_ML.

  • Regular paper track: This track will accept submissions of maximum 6 pages on any of the topics of the workshop. Each submission will be allowed to have additional pages for references and appendix, which the reviewers do not need to review (but may do so if they wish).

  • Short exploratory paper track: This track solicits short submissions of 4 pages or less that exhibit a novel preliminary observation relevant to the topics of the workshop. The observation may be:

• (A) An experimental observation, relevant to the topic of the workshop, that occurs reliably in multiple experimental settings. Based on the empirical results, the authors should provide a theoretical formalisation of the problem and propose a conjecture, including relevant assumptions.

• (B) A theoretical result, relevant to the topic of the workshop, that the authors are able to prove in a justifiable theoretical setting. Further, the authors should argue why they believe their setting is relevant for practical use and provide an experimental setting that they believe can show their phenomenon in practice.  

• All submissions in this track will be judged on the basis of the rigour and correctness of 

  • the experimental (A) and theoretical results (B) shown by the paper

  • the proposed conjecture (A) and experimental setting (B). 

• We welcome ongoing and unpublished work. We will also accept papers that are under review (at any venue, including ICLR 2023) at the time of submission, or that have been recently accepted for publication at a non-ML venue (i.e., any venue that is not ICML, NeurIPS, ICLR, or a similar conference or journal). Submissions published in venues for related fields (e.g., NLP, computer vision, statistics) are welcome. 

• The submission and reviewing will be handled on the openreview system.

• Submissions must be completely anonymised. The review process is double-blind.

• Accepted works will be presented in an in-person poster session with some works selected for contributed talks. It is mandatory that at least one of the authors will attend the workshop and present their paper. 

• The workshop is a non-archival venue and will not have official proceedings. Workshop submissions can be subsequently or concurrently submitted to other venues.

• Submissions and reviews will not be public. Only accepted papers will be made public.

Please contact us at trustml-unlimited-workshop-2023@googlegroups.com if you have questions regarding the workshop.